← Back to the lecture: Market Power I: Price-Searchers and How They Price

Lecture 14 Recap

Price-Searchers and How They Price

Why a seller that must cut its price to sell more stops short, how it picks its output and its price, what that costs buyers, and what free entry, closed markets, and cartels do to its profit.

Class meeting 21 · November 5, 2026 · 10-minute review

A price-searcher is a seller that must cut its price to sell more. When it charges one price to everyone, a price cut goes to every buyer, not just the new ones, so one more unit brings in less than its price. The seller makes units until the last one brings in just what it costs (marginal revenue equals marginal cost), and then charges the highest price at which those units sell. It stops short: buyers who value a unit above its cost go without, and that lost value is a deadweight loss. Whether the profit lasts depends on whether rivals can come in.

How to use this page

This recap covers the November 5 class. Class covered every slide: price-takers and price-searchers, marginal revenue, the airline’s choice of seats and fare, the gains and the deadweight loss, a student fare, free entry, a closed route, and cartels. TopHat questions 1 to 4 ran; question 5 was skipped, so its numbers appear here as the slide gives them. Read the five-step diagnostic first, then the lecture sections, then the checks. The separate Lecture 14 Transcript has the full explanation.

The five-step diagnostic

Use these steps whenever a question gives you a seller that faces a downward-sloping demand and charges one price.

  1. Find marginal revenue.
    When the price falls to sell more, add the revenue from the new units and subtract the markdown on the units already sold. Divide by the number of new units. Marginal revenue is below the price.
  2. Find the output.
    Make units as long as each one brings in more revenue than it costs. Stop where marginal revenue equals marginal cost.
  3. Find the price from demand.
    Go up from that output to the demand curve. The price is the highest one at which that output sells. It is not the height where marginal revenue meets marginal cost.
  4. Measure the gains and the loss.
    Profit = output × (price − cost a unit). Buyers’ gain (consumer surplus) is the triangle under demand and above the price. Compare with price = marginal cost: the triangle between that output and the seller’s output, under demand and above cost, is the deadweight loss.
  5. Ask what happens next.
    Can the seller charge different groups different prices, with no resale? Can rivals enter on the same terms? Is the market closed by law? Are the sellers trying to act as one?

What you should be able to do

  1. Tell a price-taker from a price-searcher, and explain what market power means.
  2. Compute marginal revenue from a price cut, and explain why it is below the price.
  3. Find a price-searcher’s output where marginal revenue equals marginal cost, and read its price from demand.
  4. Measure its profit, the buyers’ gain, and the deadweight loss, and compare them with price = marginal cost.
  5. Explain how charging a separate price to a group that cannot resell can raise profit and shrink the deadweight loss.
  6. Explain how free entry drives the price down to cost, and why a legal barrier turns the profit into a monopoly rent.
  7. Explain why each member of a cartel is tempted to cheat, and what a cartel needs to hold together.

The lecture this time

Class started with the plan for the rest of the semester (see below). The lecture then followed one example all the way through: suppose one airline flies the only nonstop route between two cities. Its demand is fare = $600 − seats a day, and each extra seat costs it $200.

Price-takers and price-searchers

To fill more seats, the airline must cut its fare

Lectures 12 and 13 studied price-takers: firms too small to move the market price. A price-taker sells as much as it wants at the market price ($9 in Lecture 13), so its marginal revenue equals the price, and it makes units up to where price equals marginal cost. A price-searcher faces a downward-sloping demand: it can sell more only by cutting its price, and it can push the price up by selling less. Economists call this market power. The airline is a price-searcher. If you want to get between the two cities without it, you must drive or find another way. At $500 it sells 100 seats; at $400, 200 seats. It charges one fare to everyone.

Marginal revenue · TopHat

A fare cut costs money on seats already sold

Cut the fare from $500 to $400. The 100 new riders bring in 100 × $400 = $40,000. But the first 100 riders now pay $100 less each, so the airline gives up $10,000 on seats it was already selling. Revenue rises by $30,000, or $300 a new seat on average. The new seats sell for $400, but they add only $300 each.

The instructor tied this to elasticity (not on the slides). If a price cut raises total revenue, demand is elastic there; if it lowers revenue, demand is inelastic; if revenue does not change, demand is unit elastic. A price-searcher never prices on the inelastic part of its demand: it could sell less, raise the price, and take in more revenue at a lower cost.

TopHat check. At $400 the airline sells 200 seats. To sell 300 it must cut the fare to $300 for every rider. On average, how much extra revenue does each of the 100 extra seats bring in? Revenue goes from $80,000 to $90,000, so the change is $10,000, and $10,000 ÷ 100 = $100. Most of the class got it. The answer is not $300, the new fare, because the first 200 riders now pay $100 less.

Marginal revenue is less than the fare

Marginal revenue is the extra revenue from one more seat. The table:

$500, 100 seats, $50,000 · $400, 200 seats, $80,000 (each new seat adds $300) · $300, 300 seats, $90,000 (each adds $100) · $200, 400 seats, $80,000 (each adds −$100).

From $300 to $200, revenue falls: the markdown on the first 300 riders is bigger than the fares from the new ones. Even if seats cost nothing, the airline would not cut the fare to $200. For a price-searcher that charges one price, marginal revenue falls faster than demand. For a price-taker the two are the same flat line.

The airline’s choice · TopHat

Each extra seat costs the airline $200

Lecture 12’s marginal cost rose with output. For simplicity, here each seat costs the same $200 to provide, and there are no other costs. Then rising cost is not why the airline stops; the only reason is the markdown on seats already sold. The rule is the same one a price-taker uses: add seats as long as each brings in more revenue than it costs, so stop where marginal revenue equals marginal cost. A student said it in class: you would not go past the point where a seat costs more than it brings in.

TopHat check. What fare does the airline charge to make the most profit? $400. From 100 to 200 seats, each seat adds $300 of revenue on average and costs $200: fly them. From 200 to 300, each adds only $100: do not. So it flies 200 seats, and the highest fare at which 200 seats sell is $400. The answer $200 is the cost of a seat, and the marginal revenue of the last seat, but at a fare of $200 the airline would sell 400 seats. At $300 it would sell 300 seats, and the last 100 would bring in less than they cost.

The airline flies 200 seats a day at $400

Marginal revenue = marginal cost picks the seats; demand picks the fare. Go up from 200 seats to the demand curve: the 200th rider would pay $400. Profit is 200 × ($400 − $200) = $40,000 a day. A price-taker at the lowest average total cost earns no profit; this seller does.

Gains and lost value · TopHat

Riders gain $20,000; the airline, $40,000

Gains from trade are the gap between what a buyer values a seat at (demand) and what it costs (marginal cost). Riders’ gain, the consumer surplus, is the triangle under demand and above the $400 fare: ½ × 200 × ($600 − $400) = $20,000. The airline’s gain, its producer surplus, is its profit: $40,000.

The 250th rider would pay $350 for a seat that costs $200. The airline does not sell it.

TopHat check. Why not? Selling it means a lower fare for every rider, so it adds less revenue than it costs. To sell seat 250 the airline must cut the fare to $350 for everyone, losing $50 on each of the 200 seats it already sells. From 200 to 250 seats each seat adds only $150 on average, less than $200. “Selling more seats would lower total revenue” is false here: revenue rises from $80,000 to $87,500 at 250 seats. “Revenue is already as large as it can be at 200 seats” is false too: revenue keeps rising up to 300 seats. (In class the instructor said these two could be true for some seller, and a student linked them to elasticity: they would hold only on the inelastic or the unit-elastic part of demand.) Most of the class got it right.

Seats worth more than they cost go unsold

Riders between seat 200 and seat 400 would each pay more than the $200 a seat costs, yet no seat flies for them. That lost value is the deadweight loss: ½ × 200 × ($400 − $200) = $20,000 a day, value nobody gets. Together, riders and the airline gain $60,000. If the fare were $200, 400 seats would fly and riders would gain $80,000, while the airline’s profit would fall to zero. So compared with a price equal to cost, the price-searcher makes too little output. The instructor added that it also earns less profit than it could: if it could turn the deadweight loss into sales, it could keep more. This result rests on one assumption: the airline charges everyone the same fare.

A student fare · TopHat

Suppose students can get their own fare

What could the airline do about the seats it leaves unsold? A student suggested student discounts. Suppose riders are of two kinds. Business travelers would pay $300 to $600 (they must be at a Monday meeting); students would pay less than $300 (they can be flexible). With a student ID check, the airline can keep the $400 fare for business travelers and give students their own fare. Charging different buyers different prices for the same good is price discrimination.

It works only if the low-fare group cannot resell. If a student could buy at $200 and resell to a business traveler at $300, the student would keep the $100, and the airline would not use the plan. Plane tickets carry your name and are checked against your ID, so they cannot be resold. Water bottles can. The instructor’s examples (not on the slides): universities ask for family income on the FAFSA partly to set different prices, and tuition cannot be resold; services such as car repair and medical care are hard to resell, so price discrimination is more common for services than for goods.

TopHat check. Students would buy 0 seats at $300, 25 at $275, 50 at $250, 75 at $225, and 100 at $200. Each seat costs $200. What student fare makes the most profit? $250. The class’s answers were more spread out on this one. Profit on student seats is 25 × $75 = $1,875 at $275, 50 × $50 = $2,500 at $250, and 75 × $25 = $1,875 at $225. In marginal terms: going from 25 to 50 student seats adds $225 a seat on average, more than $200; going from 50 to 75 adds only $175. A $200 fare fills 100 seats, but the last ones bring in less than they cost.

Each group’s last seat brings in $200

Price discrimination is like facing two demand curves with the same $200 cost. For each group the airline makes seats until marginal revenue equals marginal cost, then reads that group’s fare from its own demand: 200 business seats at $400 and 50 student seats at $250. The fares differ because students are more sensitive to price. Different fares are the result, not the goal.

New student seats shrink the deadweight loss

The airline now sells 250 seats. It gains $50 on each of 50 student seats, $2,500 a day (so $42,500 in all); students gain ½ × 50 × $50 = $1,250; business travelers are no worse off. The deadweight loss falls from $20,000 to $15,000 + $1,250 = $16,250. The $15,000 is the business travelers who still do not fly; the $1,250 is the students past seat 50. Here the loss falls because every student seat is a new sale. A price gap that only moves riders from one fare to another adds no value. In class the instructor said a ban on price discrimination would make the deadweight loss bigger and make us poorer. (He said “legal”; he meant a ban.)

Perfect price discrimination (not on the slides)

Suppose the seller knew each buyer’s exact willingness to pay and charged it. Then every rider who values a seat above $200 flies, so there is no deadweight loss, but there is also no consumer surplus: the whole triangle becomes producer surplus. On the route, riders lose their $20,000, and the airline’s profit rises from $40,000 to $80,000: it takes the riders’ $20,000 and also the $20,000 that was deadweight loss. (In class the instructor said producers gain what consumers lose; that holds for each buyer who was already buying.) The instructor’s view: the seller’s owners are people too, and anyone with a retirement account likely owns part of an airline, so a rise in profit raises their wealth. Whether society is worse off depends on how much weight you put on who gets the gain.

Two conditions again: the good cannot easily be resold, and the seller has market power. A gas station does not charge a UPS driver more than other drivers: the driver could cross the street, or ask someone to fill a gas can at the lower price. A student asked about prices that depend on your location. The instructor’s answer: the more a seller learns about you, the easier it is to price discriminate, which helps buyers with a low willingness to pay, who might not buy at all otherwise. A buyer with a high willingness to pay should not reveal it, as when shopping for a car.

Free entry, closed routes, and cartels

Rivals’ seats push the fare down

Back to one fare for everyone. Suppose any airline may fly the route on the same terms, at the same $200 a seat. The $40,000 a day draws rivals. The instructor’s illustration (not on the slides): a new airline charges $399, the first airline answers with $398, the newcomer with $397, and so on. More seats sell only at a lower fare. At 300 seats the fare is $300, and the airlines together earn 300 × ($300 − $200) = $30,000 a day, so still more come in.

Entry stops when the fare reaches $200

As in Lecture 13, entry goes on while there is profit to win. It stops when the fare equals the $200 cost of a seat: 400 seats fly, the airlines earn zero economic profit (an ordinary return), riders gain $80,000 a day, and the $20,000 deadweight loss is gone. So a single seller today does not prove lasting market power. The instructor’s example: Apple’s iPhone soon faced Samsung’s Galaxy, and competition pushed prices down.

The $40,000 lasts if rivals are kept out

A better test of monopoly is whether a law or another barrier keeps rivals out. The FAA limits scheduled takeoffs and landings at JFK, LaGuardia, and Reagan National: an airline needs a slot for each. Suppose slots keep rivals off our route. Then no new seats push the fare down, and the $40,000 can last. It is a monopoly rent: it exists because entry is blocked, and, like Lecture 13’s taxi medallions, it goes to whoever holds the slots. If Apple had persuaded Congress to ban every phone but the iPhone, its profit would not have been competed away. Not every lasting profit is a monopoly rent: it can pay for a new idea, a risk, or a scarce input, and a fare rise caused by higher fuel costs is not market power.

Suppose four airlines agree to act as one

Four airlines fly the route at one common fare, each at $200 a seat. They form a cartel: they agree to act as one airline. Together they fly 200 seats a day at $400, 50 seats each, and each earns 50 × ($400 − $200) = $10,000 a day, a quarter of the one airline’s $40,000. The instructor noted that drug cartels work the same way: they hold back output to push the price up. (In class he called “oligopoly” another name for a cartel. An oligopoly is a market with only a few sellers; a cartel is an agreement among sellers to act as one.)

Skipped TopHat question, from the slide. Airline A breaks the deal and flies 100 seats while the others keep to 50. The route now has 250 seats, which sell only at $350. A earns 100 × $150 = $15,000; each of the others earns 50 × $150 = $7,500. Every member is tempted the same way. If all four fly 100 seats, 400 seats sell at $200, and each earns $0.

Each member is tempted to cheat

Each member wants the others to keep the deal while it sells more. Members cannot easily see what the others do, and price-fixing deals are illegal under the Sherman Antitrust Act, so a member cannot ask the courts to enforce one. To hold, a cartel must find every seller, keep new ones out, and catch cheaters. Without an outside enforcer, cartels tend to break down. That is one reason sellers sometimes ask to be regulated. The instructor’s example: the Civil Aeronautics Board once decided which airlines could fly which routes, and he said airfares were then far higher than they are now. Lecture 15 takes up antitrust and regulation.

Announcements (as given in class)

No class Tuesday, November 10. A Canvas announcement explains it. Class meets again Thursday, November 12, and twice the week after. There is no class in Thanksgiving week. After Thanksgiving come a final exam review and the final exam. Counting November 5, four classes of new material were left.

The final exam has 30 questions, not 40, so you have a little more time a question: 10 on the material of the first exam, 10 on the material of the second exam, and 10 on the material since the second exam.

At the end of class the instructor noted that the university was closing soon that evening because ESPN was coming for the football game.

Connect each example to its lesson

Do not memorize an example as a story. Use it to recover the economic principle.

Examples are memory cues; the right column is the principle each example should help you recover.
Example from class Economic lesson
$500 to $400: +$40,000 of new seats, −$10,000 of markdown A price cut goes to every buyer, so marginal revenue is below the price.
$300 to $200: revenue falls from $90,000 to $80,000 Past the point where revenue peaks, marginal revenue is negative; a seller never prices there.
200 seats at $400, profit $40,000 Marginal revenue = marginal cost picks the output; demand picks the price.
The 250th rider would pay $350 for a $200 seat A single-price seller stops short, and value that exceeds cost goes unsold: the deadweight loss.
A $250 student fare with an ID check; the FAFSA Price discrimination needs groups that can be told apart and cannot resell; it can shrink the deadweight loss.
Charging every rider exactly what they would pay Perfect price discrimination removes the deadweight loss and turns all consumer surplus into profit.
$399, $398, $397; Apple and Samsung With free entry, rivals push the price down to cost and the profit disappears.
FAA slots at JFK, LaGuardia, and Reagan National When the law blocks entry, the profit lasts as a monopoly rent for whoever holds the right to enter.
Four airlines at $10,000 each; the Civil Aeronautics Board Each cartel member gains by cheating, so cartels need an outside enforcer to hold.

Check your reasoning

These use new numbers. Questions 1 to 4 and 7 to 10 use one example: a ferry is the only boat on its route and charges one fare to everyone. Its demand is fare = $80 − riders a day (at $70 it carries 10 riders, at $60 20, at $50 30, at $40 40, at $30 50). Each extra rider costs it $20. Try each question before you open the answer.

Question 1 — The ferry cuts its fare from $60 to $50, and its riders go from 20 to 30. On average, how much extra revenue does each of the 10 extra riders bring in?

The new riders bring in 10 × $50 = $500. The first 20 riders now pay $10 less, so the ferry gives up 20 × $10 = $200. Revenue rises by $300 (from $1,200 to $1,500), or $30 a new rider. That is less than the $50 fare.

Question 2 — How many riders does the ferry carry, what fare does it charge, and what is its profit?

From 20 to 30 riders, each adds $30 of revenue on average, more than the $20 cost: carry them. From 30 to 40, each adds only $10 ($1,500 to $1,600): do not. So it carries 30 riders, where marginal revenue = $80 − 2 × 30 = $20 = marginal cost. The fare comes from demand: $50. Profit = 30 × ($50 − $20) = $900 a day. The answer $20 is the cost, not the fare.

Question 3 — Measure the riders’ gain and the deadweight loss. What would riders gain if the fare equaled the $20 cost?

Riders’ gain = ½ × 30 × ($80 − $50) = $450. At a $20 fare, 60 riders would go. The deadweight loss is the triangle between 30 and 60 riders, under demand and above cost: ½ × 30 × ($50 − $20) = $450. At a $20 fare riders would gain ½ × 60 × ($80 − $20) = $1,800, which is $450 + $900 + $450.

Question 4 — A 35th rider would pay $45, and carrying that rider costs only $20. Why does the ferry not carry riders 31 to 35?

To carry 35 riders it must cut the fare to $45 for everyone. Revenue goes from $1,500 to 35 × $45 = $1,575, up only $75, or $15 a rider, less than the $20 cost. Profit would fall from $900 to 35 × $25 = $875. The rider’s $45 is above cost, but the markdown on the first 30 riders makes the sale unprofitable.

Question 5 — A different ferry carries tourists. Local residents, who show an ID, would ride 0 times a day at $40, 10 at $35, 20 at $30, 30 at $25, and 40 at $20. Each rider costs $20, and the ID check keeps the local fare separate. What local fare makes the most profit, and how much do locals gain?

Profit on local riders: 10 × $15 = $150 at $35, 20 × $10 = $200 at $30, 30 × $5 = $150 at $25, and $0 at $20. So the local fare is $30 for 20 riders. In marginal terms, going from 10 to 20 local riders adds $25 a rider on average, more than $20; going from 20 to 30 adds only $15. Locals gain ½ × 20 × ($40 − $30) = $100.

Question 6 — Why does the plan in Question 5 fail if locals can buy tickets and hand them to tourists?

Locals would buy at $30 and resell to tourists for less than the tourist fare, keeping the difference. Tourists would stop buying at the high fare, and the two fares could not stay apart. Price discrimination needs groups that can be told apart and a good that cannot be resold.

Question 7 — Back to the first ferry. Any boat may run the route at the same $20 a rider. Where does the fare end up, and what happens to profit and the deadweight loss?

The $900 profit draws rivals, and more riders sell only at lower fares. At 40 riders the fare is $40, and the boats together still earn 40 × $20 = $800, so more come in. Entry stops at a fare of $20, with 60 riders: zero economic profit, riders gain $1,800, and the deadweight loss is gone.

Question 8 — Instead, the city issues one ferry license for the route and lets no one else run. Who gets the $900 a day? What does a buyer of the license later earn?

The $900 a day is a monopoly rent: it exists because entry is blocked. It goes to whoever holds the license. If the license can be sold, a buyer pays today for that whole stream of $900 a day, so the buyer earns only an ordinary return on what was paid. Riders keep paying $50.

Question 9 — Three ferries form a cartel on the route: 30 riders in all at $50, 10 riders each. Ferry A breaks the deal and carries 20 riders while the others keep to 10. What fare do riders pay, and what does each ferry earn?

Under the deal, each earns 10 × ($50 − $20) = $300 a day. With A at 20, the route carries 40 riders, which sell only at $40. A earns 20 × $20 = $400; each of the others earns 10 × $20 = $200. A gains by cheating while the others lose. If all three carry 20, 60 riders sell at $20 and each earns $0.

Question 10 — Suppose the first ferry knew each rider’s exact willingness to pay and charged it. How many riders go? What are the riders’ gain, the ferry’s profit, and the deadweight loss?

Every rider who values a trip above the $20 cost goes: 60 riders. Each pays exactly what the trip is worth to them, so riders gain $0. The ferry’s profit is the whole triangle, ½ × 60 × ($80 − $20) = $1,800, and the deadweight loss is $0.

Bottom line

A seller that must cut its price to sell more takes a lower price on every unit, so one more unit brings in less than its price. It makes units until marginal revenue equals marginal cost, charges what demand will bear for that output, and leaves sales worth more than they cost unmade. Separate prices for groups that cannot resell can recover some of that lost value. Free entry competes the profit away; a legal barrier lets it last as a monopoly rent; and a cartel that tries to act as one seller is undone by each member’s reason to cheat.

← Back to the lecture: Market Power I: Price-Searchers and How They Price