This preview shows page 1. Sign up to view the full content.
Unformatted text preview: 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 287 13 Oligopoly After studying this chapter,
y ou will be able to:
■ Define and identify oligopoly ■ Explain two traditional oligopoly models ■ Use game theory to explain how price and output are
determined in oligopoly ■ Use game theory to explain other strategic decisions ■ Describe the antitrust laws that regulate oligopoly A n intense price war in the market for PCs has The theories of perfect competition and monopoly don’t pre- driven the price of a laptop below $1,000 and the price of a dict the behavior of the firms we’ve just described. To under- desktop below $500. A handful of firms—Dell, Hewlett- stand how markets work when only a handful of firms compete, Packard, Lenovo, Acer, and Toshiba—account for more than we need the richer models that are explained in this chapter. In one half of the global market. Each of these firms must pay Reading Between the Lines at the end of this chapter, we’ll return close attention to what the other firms are doing to the market for personal computers and see how Dell and In some markets, there are only two firms. Computer chips Hewlett-Packard slugged it out for dominance in that market. are an example. The chips that drive most PCs are made by
Intel and Advanced Micro Devices. How does competition
between just two chip makers work?
When a small number of firms compete in a market, do they
operate in the social interest, like firms in perfect competition?
Or do they restrict output to increase profit, like a monopoly? 287 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 288 CHAPTER 13 Oligopoly 288 ◆ What Is Oligopoly? total cost curve of a taxi company is ATC1 in part (a),
the market is a natural duopoly—an oligopoly market
with two firms. You can probably see some examples
of duopoly where you live. Some cities have only two
taxi companies, two car rental firms, two copy centers, or two college bookstores.
The lowest price at which the firm would remain
in business is $10 a ride. At that price, the quantity
of rides demanded is 60 a day, the quantity that can
be provided by just two firms. There is no room in
this market for three firms. But if there were only one
firm, it would make an economic profit and a second
firm would enter to take some of the business and
If the average total cost curve of a taxi company is
ATC2 in part (b), the efficient scale of one firm is 20 rides
a day. This market is large enough for three firms.
A legal oligopoly arises when a legal barrier to entry
protects the small number of firms in a market. A city
might license two taxi firms or two bus companies, for
example, even though the combination of demand and
economies of scale leaves room for more than two firms. Oligopoly, like monopolistic competition, lies
between perfect competition and monopoly. The
firms in oligopoly might produce an identical product and compete only on price, or they might produce a differentiated product and compete on price,
product quality, and marketing. Oligopoly is a market structure in which
■ Natural or legal barriers prevent the entry of new
■ A small number of firms compete. Barriers to Entry
Natural or legal barriers to entry can create oligopoly. You saw in Chapter 11 how economies of scale
and demand form a natural barrier to entry that can
create a natural monopoly. These same factors can
create a natural oligopoly.
Figure 13.1 illustrates two natural oligopolies. The
demand curve, D (in both parts of the figure), shows
the demand for taxi rides in a town. If the average Price and cost (dollars per ride) Price and cost (dollars per ride) Natural Oligopoly FIGURE 13.1 25 20 ATC1
15 10 5 0 25 20 ATC2
ATC D D
30 60 90 Quantity (rides per day)
(a) Natural duopoly The lowest possible price is $10 a ride, which is the minimum
average total cost. When a firm produces 30 rides a day, the
efficient scale, two firms can satisfy the market demand. This
natural oligopoly has two firms—a natural duopoly.
one firm 5 0 20 Three firms
40 60 80 Quantity (rides per day)
(b) Natural oligopoly with three firms When the efficient scale of one firm is 20 rides per day,
three firms can satisfy the market demand at the lowest
possible price. This natural oligopoly has three firms. 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 289 What Is Oligopoly? 289 Small Number of Firms Examples of Oligopoly Because barriers to entry exist, oligopoly consists of
a small number of firms, each of which has a large
share of the market. Such firms are interdependent,
and they face a temptation to cooperate to increase
their joint economic profit. The box below shows some examples of oligopoly.
The dividing line between oligopoly and monopolistic competition is hard to pin down. As a practical
matter, we identify oligopoly by looking at concentration ratios, the Herfindahl-Hirschman Index, and
information about the geographical scope of the
market and barriers to entry. The HHI that divides
oligopoly from monopolistic competition is generally taken to be 1,000. An HHI below 1,000 is usually an example of monopolistic competition, and a
market in which the HHI exceeds 1,000 is usually
an example of oligopoly. Interdependence With a small number of firms in a market, each firm’s actions influence the profits of all
the other firms. When Penny Stafford opened her coffee shop in Bellevue, Washington, a nearby Starbucks
coffee shop took a hit. Within days, Starbucks began
to attract Penny’s customers with enticing offers and
lower prices. Starbucks survived but Penny eventually
went out of business. Penny Stafford and Starbucks
Temptation to Cooperate When a small number of
firms share a market, they can increase their profits
by forming a cartel and acting like a monopoly. A
cartel is a group of firms acting together—colluding—to limit output, raise price, and increase economic profit. Cartels are illegal, but they do operate
in some markets. But for reasons that you’ll discover
in this chapter, cartels tend to break down. Oligopoly Today
Near Duopoly in Batteries
These markets are oligopolies.
Although in some of them, the number of firms (in parentheses) is large,
the share of the market held by the 4
largest firms (the red bars) is close to
The most concentrated
markets—cigarettes, glass bottles and
jars, washing machines and dryers,
and batteries, are dominated by just
one or two firms.
If you want to buy an AAA battery
for your TV remote or toothbrush,
you’ll find it hard to avoid buying a
Duracell or an Energizer. Review Quiz ◆
4 What are the two distinguishing characteristics
Why are firms in oligopoly interdependent?
Why do firms in oligopoly face a temptation to
Can you think of some examples of oligopolies
that you buy from?
Work Study Plan 13.1
and get instant feedback. Industry (number of firms) Herfindahl-Hirschman Index Cigarettes (9) – Glass bottles and jars (11) 2,960 Washing machines and dryers (10) 2,870 Batteries (35) 2,883 Light bulbs (54) 2,849 Breakfast cereals (48) 2,446 House slippers (22) 2,053 Automobiles (173) 2,350 Chocolate (152) 2,567 Motorcycles (373) 2,037
0 20 40 60 80 100 Percentage of industry total revenue
4 largest firms
Measures of Concentration Source of data: U.S. Census Bureau. next 4 largest firms next 12 largest firms 9160335_CH13_p287-314.qxd 9:05 AM Page 290 CHAPTER 13 Oligopoly ◆ Two Traditional Oligopoly Models
Suppose you run one of three gas stations in a small
town. You’re trying to decide whether to cut your
price. To make your decision, you must predict how
the other firms will react and calculate the effects of
those reactions on your profit. If you cut your price
and your competitors don’t cut theirs, you sell more
and the other two firms sell less. But won’t the other
firms cut their prices too and make your profits fall?
What will you do?
Several models have been developed to explain the
prices and quantities in oligopoly markets. The models
fall into two broad groups: traditional models and
game theory models. We’ll look at examples of both
types, starting with two traditional models. The Kinked Demand Curve Model
The kinked demand curve model of oligopoly is
based on the assumption that each firm believes that
if it raises its price, others will not follow, but if it
cuts its price, other firms will cut theirs.
Figure 13.2 shows the demand curve (D) that a
firm believes it faces. The demand curve has a kink
at the current price, P, and quantity, Q. At prices
above P, a small price rise brings a big decrease in
the quantity sold. If one firm raises its price, other
firms will hold their current price constant. The
firm that raised its price will have the highest price
and it will lose market share. At prices below P,
even a large price cut brings only a small increase in
the quantity sold. In this case, if one firm cuts its
price, other firms will match the price cut. The firm
that cuts its price will get no price advantage over its
The kink in the demand curve creates a break in the
marginal revenue curve (MR). To maximize profit, the
firm produces the quantity at which marginal cost
equals marginal revenue. That quantity, Q, is where the
marginal cost curve passes through the gap AB in the
marginal revenue curve. If marginal cost fluctuates
between A and B, like the marginal cost curves MC0 and
MC1, the firm does not change its price or its output.
Only if marginal cost fluctuates outside the range AB
does the firm change its price and output. So the kinked
demand curve model predicts that price and quantity
are insensitive to small cost changes.
But this model has a problem. If marginal cost
increases by enough to cause the firm to increase its
price and if all firms experience the same increase in FIGURE 13.2
Price and cost (dollars) 290 6/22/09 The Kinked Demand
Curve Model MC
P 1 MC 0
A B MR
0 Q D
Quantity The price in an oligopoly market is P. Each firm believes it
faces the demand curve D. At prices above P, a small price
rise brings a big decrease in the quantity sold because
other firms do not raise their prices. At prices below P, even
a big price cut brings only a small increase in the quantity
sold because other firms also cut their prices. Because the
demand curve is kinked, the marginal revenue curve, MR,
has a break AB. Profit is maximized by producing Q. The
marginal cost curve passes through the break in the marginal revenue curve. Changes in marginal cost inside the
range AB leave the price and quantity unchanged.
animation marginal cost, they all increase their prices together. The
firm’s belief that others will not join it in a price rise is
incorrect. A firm that bases its actions on beliefs that are
wrong does not maximize profit and might even end up
incurring an economic loss. Dominant Firm Oligopoly
A second traditional model explains a dominant
firm oligopoly, which arises when one firm—the
dominant firm—has a big cost advantage over the
other firms and produces a large part of the industry
output. The dominant firm sets the market price
and the other firms are price takers. Examples of
dominant firm oligopoly are a large gasoline retailer
or a big box store that dominates its local market. 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 291 Two Traditional Oligopoly Models A Dominant Firm Oligopoly
S Price and cost (dollars per gallon) Price (dollars per gallon) FIGURE 13.3 10 4 3 291 A B 4 MC 3 A B D
2 XD 2 MR
Quantity (thousands of gallons per week) (a) Ten small firms and market demand The demand curve for gas in a city is D in part (a). There are
10 small competitive firms that together have a supply curve
of S10. In addition, there is 1 large firm, Big-G, shown in part
(b). Big-G faces the demand curve XD, determined as the market demand D minus the supply of the 10 small firms S10 —the
demand that is not satisfied by the small firms. Big-G’s marginal 0 10
Quantity (thousands of gallons per week) (b) Big-G‘s price and output decision revenue curve is MR and its marginal cost curve is MC. Big-G
sets its output to maximize profit by equating marginal cost
and marginal revenue. This output is 10,000 gallons per
week. The price at which Big-G can sell this quantity is $3 a
gallon. The 10 small firms take this price, and each firm sells
1,000 gallons per week, point A in part (a). animation To see how a dominant firm oligopoly works, suppose that 11 firms operate gas stations in a city. BigG is the dominant firm. Figure 13.3 shows the
market for gas in this city. In part (a), the demand
curve D tells us the total quantity of gas demanded in
the city at each price. The supply curve S10 is the supply curve of the 10 small firms. Part (b) shows the situation facing Big-G. Its marginal cost curve is MC.
Big-G faces the demand curve XD, and its marginal
revenue curve is MR. The demand curve XD shows
the excess demand not met by the 10 small firms. For
example, at a price of $3 a gallon, the quantity
demanded is 20,000 gallons, the quantity supplied by
the 10 small firms is 10,000 gallons, and the excess
quantity demanded is 10,000 gallons, measured by
the distance AB in both parts of the figure.
To maximize profit, Big-G operates like a monopoly. It sells 10,000 gallons a week, where its marginal
revenue equals its marginal cost, for a price of $3 a
gallon. The 10 small firms take the price of $3 a gallon. They behave just like firms in perfect competition.
The quantity of gas demanded in the entire city at $3
a gallon is 20,000 gallons, as shown in part (a). Of this
amount, Big-G sells 10,000 gallons and the 10 small
firms each sell 1,000 gallons. Review Quiz ◆
1 2 What does the kinked demand curve model
predict and why must it sometimes make a prediction that contradicts its basic assumption?
Do you think a market with a dominant firm is
in long-run equilibrium? Explain why or why not.
Work Study Plan 13.2
and get instant feedback. The traditional models don’t enable us to understand all oligopoly markets and we’re now going to
study some newer models based on game theory. 9160335_CH13_p287-314.qxd 292 6/22/09 9:05 AM Page 292 CHAPTER 13 Oligopoly ◆ Oligopoly Games
■ Economists think about oligopoly as a game, and to
study oligopoly markets they use a set of tools called
game theory. Game theory is a tool for studying
strategic behavior—behavior that takes into account
the expected behavior of others and the recognition
of mutual interdependence. Game theory was
invented by John von Neumann in 1937 and
extended by von Neumann and Oskar Morgenstern
in 1944. Today, it is one of the major research fields
Game theory seeks to understand oligopoly as
well as other forms of economic, political, social, and
even biological rivalries by using a method of analysis
specifically designed to understand games of all types,
including the familiar games of everyday life. We
will begin our study of game theory and its application to the behavior of firms by thinking about
familiar games. What Is a Game?
What is a game? At first thought, the question
seems silly. After all, there are many different games.
There are ball games and parlor games, games of
chance and games of skill. But what is it about all
these different activities that make them games?
What do all these games have in common? We’re
going to answer these questions by looking at a
game called “the prisoners’ dilemma.” This game
captures the essential features of many games,
including oligopoly, and it gives a good illustration
of how game theory works and how it generates
predictions. The Prisoners’ Dilemma
Art and Bob have been caught red-handed stealing a
car. Facing airtight cases, they will receive a sentence
of two years each for their crime. During his interviews with the two prisoners, the district attorney
begins to suspect that he has stumbled on the two
people who were responsible for a multimilliondollar bank robbery some months earlier. But this
is just a suspicion. He has no evidence on which he
can convict them of the greater crime unless he can
get them to confess. But how can he extract a confession? The answer is by making the prisoners play
a game. The district attorney makes the prisoners
play the following game. ■
■ All games share four common features:
Outcome Rules Each prisoner (player) is placed in a separate
room and cannot communicate with the other
prisoner. Each is told that he is suspected of having
carried out the bank robbery and that
If both of them confess to the larger crime, each will
receive a sentence of 3 years for both crimes.
If he alone confesses and his accomplice does not, he
will receive only a 1-year sentence while his accomplice will receive a 10-year sentence. Strategies In game theory, strategies are all the possible actions of each player. Art and Bob each have two
possible actions: 1. Confess to the bank robbery.
2. Deny having committed the bank robbery.
Because there are two players, each with two strategies, there are four possible outcomes:
1. Both confess.
2. Both deny.
3. Art confesses and Bob denies.
4. Bob confesses and Art denies.
Payoffs Each prisoner can work out his payoff in
each of these situations, and we can tabulate the four
possible payoffs for each of the prisoners in what is
called a payoff matrix for the game. A payoff matrix is
a table that shows the payoffs for every possible
action by each player for every possible action by
each other player.
Table 13.1 shows a payoff matrix for Art and Bob.
The squares show the payoffs for each prisoner—the
red triangle in each square shows Art’s and the blue
triangle shows Bob’s. If both prisoners confess (top
left), each gets a prison term of 3 years. If Bob confesses but Art denies (top right), Art gets a 10-year
sentence and Bob gets a 1-year sentence. If Art confesses and Bob denies (bottom left), Art gets a 1-year
sentence and Bob gets a 10-year sentence. Finally, if
both of them deny (bottom right), neither can be
convicted of the bank robbery charge but both are
sentenced for the car theft—a 2-year sentence. 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 293 Oligopoly Games Outcome The choices of both players determine the
outcome of the game. To predict that outcome, we
use an equilibrium idea proposed by John Nash of
Princeton University (who received the Nobel Prize
for Economic Science in 1994 and was the subject of
the 2001 movie A Beautiful Mind ). In Nash equilibrium,
player A takes the best possible action given the
action of player B and player B takes the best possible
action given the action of player A.
In the case of the prisoners’ dilemma, the Nash
equilibrium occurs when Art makes his best choice
given Bob’s choice and when Bob makes his best
choice given Art’s choice.
To find the Nash equilibrium, we compare all the
possible outcomes associated with each choice and
eliminate those that are dominated—that are not as
good as some other choice. Let’s find the Nash equilibrium for the prisoners’ dilemma game.
Finding the Nash Equilibrium Look at the situation from Art’s point of view. If Bob confesses (top row),
Art’s best action is to confess because in that case, he
is sentenced to 3 years rather than 10 years. If Bob
denies (bottom row), Art’s best action is still to confess because in that case he receives 1 year rather than
2 years. So Art’s best action is to confess.
Now look at the situation from Bob’s point of
view. If Art confesses (left column), Bob’s best action
is to confess because in that case, he is sentenced to
3 years rather than 10 years. If Art denies (right column), Bob’s best action is still to confess because in
that case, he receives 1 year rather than 2 years. So
Bob’s best action is to confess.
Because each player’s best action is to confess, each
does confess, each goes to jail for 3 years, and the district attorney has solved the bank robbery. This is the
Nash equilibrium of the game.
The Dilemma Now that you have found the outcome
to the prisoners’ dilemma, you can better see the
dilemma. The dilemma arises as each prisoner contemplates the consequences of denying. Each prisoner
knows that if both of them deny, they will receive
only a 2-year sentence for stealing the car. But neither
has any way of knowing that his accomplice will deny.
Each poses the following questions: Should I deny
and rely on my accomplice to deny so that we will
both get only 2 years? Or should I confess in the hope
of getting just 1 year (provided that my accomplice
denies) knowing that if my accomplice does confess, TABLE 13.1 293 Prisoners’ Dilemma
Confess 3 years Deny 10 years Confess
3 years 1 year Bob's
1 year 2 years Deny
10 years 2 years Each square shows the payoffs for the two players, Art
and Bob, for each possible pair of actions. In each
square, the red triangle shows Art’s payoff and the blue
triangle shows Bob’s. For example, if both confess, the
payoffs are in the top left square. The equilibrium of the
game is for both players to confess and each gets a 3year sentence. we will both get 3 years in prison? The dilemma leads
to the equilibrium of the game.
A Bad Outcome For the prisoners, the equilibrium of the game, with each confessing, is not the best outcome. If neither of them confesses, each gets only 2
years for the lesser crime. Isn’t there some way in
which this better outcome can be achieved? It seems
that there is not, because the players cannot communicate with each other. Each player can put himself in
the other player’s place, and so each player can figure
out that there is a best strategy for each of them. The
prisoners are indeed in a dilemma. Each knows that
he can serve 2 years only if he can trust the other to
deny. But each prisoner also knows that it is not in
the best interest of the other to deny. So each prisoner knows that he must confess, thereby delivering a
bad outcome for both.
The firms in an oligopoly are in a similar situation
to Art and Bob in the prisoners’ dilemma game. Let’s
see how we can use this game to understand oligopoly. 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 294 CHAPTER 13 Oligopoly 294 An Oligopoly Price-Fixing Game
We can use game theory and a game like the prisoners’ dilemma to understand price fixing, price wars,
and other aspects of the behavior of firms in oligopoly. We’ll begin with a price-fixing game.
To understand price fixing, we’re going to study
the special case of duopoly—an oligopoly with two
firms. Duopoly is easier to study than oligopoly with
three or more firms, and it captures the essence of all
oligopoly situations. Somehow, the two firms must
share the market. And how they share it depends on
the actions of each. We’re going to describe the costs
of the two firms and the market demand for the item
they produce. We’re then going to see how game theory helps us to predict the prices charged and the
quantities produced by the two firms in a duopoly.
Cost and Demand Conditions Two firms, Trick and MC ATC 6 Minimum
ATC 0 1 into a collusive agreement. A collusive agreement is an
agreement between two (or more) producers to form
a cartel to restrict output, raise the price, and increase
profits. Such an agreement is illegal in the United
States and is undertaken in secret.The strategies that
firms in a cartel can pursue are to
Cheat A firm that complies carries out the agreement. A
firm that cheats breaks the agreement to its own
benefit and to the cost of the other firm.
Because each firm has two strategies, there are four
possible combinations of actions for the firms:
1. Both firms comply.
2. Both firms cheat.
3. Trick complies and Gear cheats.
4. Gear complies and Trick cheats. Costs and Demand 2
Quantity (thousands of
switchgears per week) (a) Individual firm animation Price (thousands of dollars per unit) Price and cost
(thousands of dollars per unit) 10 Collusion We’ll suppose that Trick and Gear enter ■ Gear, produce switchgears. They have identical costs.
Figure 13.4(a) shows their average total cost curve
(ATC ) and marginal cost curve (MC ). Figure 13.4(b)
shows the market demand curve for switchgears (D).
The two firms produce identical switchgears, so one
firm’s switchgear is a perfect substitute for the other’s,
and the market price of each firm’s product is identical. The quantity demanded depends on that price—
the higher the price, the smaller is the quantity
demanded. FIGURE 13.4 This industry is a natural duopoly. Two firms can
produce this good at a lower cost than either one firm
or three firms can. For each firm, average total cost is
at its minimum when production is 3,000 units a
week. When price equals minimum average total
cost, the total quantity demanded is 6,000 units a
week, and two firms can just produce that quantity. 10 6 D 0 1 (b) Industry 2 3
Quantity (thousands of
switchgears per week) The average total cost curve for each firm
is ATC, and the marginal cost curve is
MC (part a). Minimum average total cost
is $6,000 a unit, and it occurs at a production of 3,000 units a week.
Part (b) shows the market demand
curve. At a price of $6,000, the quantity
demanded is 6,000 units per week. The
two firms can produce this output at the
lowest possible average cost. If the market had one firm, it would be profitable
for another to enter. If the market had
three firms, one would exit. There is room
for only two firms in this industry. It is a
natural duopoly. 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 295 O ligopoly Games Colluding to Maximize Profits Let’s work out the
payoffs to the two firms if they collude to make the
maximum profit for the cartel by acting like a
monopoly. The calculations that the two firms perform are the same calculations that a monopoly performs. (You can refresh your memory of these
calculations by looking at Chapter 11, pp. 250–251.)
The only thing that the firms in duopoly must do
beyond what a monopoly does is to agree on how
much of the total output each of them will produce.
Figure 13.5 shows the price and quantity that
maximize industry profit for the duopoly. Part (a)
shows the situation for each firm, and part (b) shows
the situation for the industry as a whole. The curve
labeled MR is the industry marginal revenue curve.
This marginal revenue curve is like that of a singleprice monopoly (Chapter 11, p. 248).The curve
labeled MCI is the industry marginal cost curve if
each firm produces the same quantity of output. This
curve is constructed by adding together the outputs
of the two firms at each level of marginal cost.
Because the two firms are the same size, at each level
of marginal cost, the industry output is twice the output of one firm. The curve MCI in part (b) is twice as
far to the right as the curve MC in part (a).
To maximize industry profit, the firms in the
duopoly agree to restrict output to the rate that
makes the industry marginal cost and marginal revenue equal. That output rate, as shown in part (b), is
4,000 units a week. The demand curve shows that the 10 MC ATC 9
6 highest price for which the 4,000 switchgears can be
sold is $9,000 each. Trick and Gear agree to charge
To hold the price at $9,000 a unit, production
must be 4,000 units a week. So Trick and Gear must
agree on output rates for each of them that total
4,000 units a week. Let’s suppose that they agree to
split the market equally so that each firm produces
2,000 switchgears a week. Because the firms are identical, this division is the most likely.
The average total cost (ATC ) of producing 2,000
switchgears a week is $8,000, so the profit per unit is
$1,000 and economic profit is $2 million (2,000 units
$1,000 per unit). The economic profit of each firm is
represented by the blue rectangle in Fig. 13.5(a).
We have just described one possible outcome for a
duopoly game: The two firms collude to produce the
monopoly profit-maximizing output and divide that
output equally between themselves. From the industry point of view, this solution is identical to a
monopoly. A duopoly that operates in this way is
indistinguishable from a monopoly. The economic
profit that is made by a monopoly is the maximum
total profit that can be made by the duopoly when
the firms collude.
But with price greater than marginal cost, either
firm might think of trying to increase profit by cheating on the agreement and producing more than the
agreed amount. Let’s see what happens if one of the
firms does cheat in this way. Colluding to Make Monopoly Profits Economic
profit Price and cost
(thousands of dollars per unit) Price and cost
(thousands of dollars per unit) FIGURE 13.5 10
9 Collusion achieves
monopoly outcome MCI 6 D MR
0 1 295 2
Quantity (thousands of
switchgears per week) (a) Individual firm 0 (b) Industry animation 1 2 4
Quantity (thousands of
switchgears per week) The industry marginal cost curve, MCI in
part (b), is the horizontal sum of the two
firms’ marginal cost curves, MC in part
(a). The industry marginal revenue curve is
MR. To maximize profit, the firms produce
4,000 units a week (the quantity at which
marginal revenue equals marginal cost).
They sell that output for $9,000 a unit.
Each firm produces 2,000 units a week.
Average total cost is $8,000 a unit, so
each firm makes an economic profit of
$2 million (blue rectangle)—2,000 units
multiplied by $1,000 profit a unit. 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 296 CHAPTER 13 Oligopoly One Firm Cheats on a Collusive Agreement To set
the stage for cheating on their agreement, Trick convinces Gear that demand has decreased and that it
cannot sell 2,000 units a week. Trick tells Gear that it
plans to cut its price so that it can sell the agreed
2,000 units each week. Because the two firms produce an identical product, Gear matches Trick’s price
cut but still produces only 2,000 units a week.
In fact, there has been no decrease in demand.
Trick plans to increase output, which it knows will
lower the price, and Trick wants to ensure that Gear’s
output remains at the agreed level.
Figure 13.6 illustrates the consequences of Trick’s
cheating. Part (a) shows Gear (the complier); part (b)
shows Trick (the cheat); and part (c) shows the industry as a whole. Suppose that Trick increases output to
3,000 units a week. If Gear sticks to the agreement to
produce only 2,000 units a week, total output is now
5,000 a week, and given demand in part (c), the price
falls to $7,500 a unit.
Gear continues to produce 2,000 units a week at a
cost of $8,000 a unit and incurs a loss of $500 a unit,
or $1 million a week. This economic loss is shown by
the red rectangle in part (a). Trick produces 3,000
units a week at an average total cost of $6,000 each.
With a price of $7,500, Trick makes a profit of One Firm Cheats
loss 0 1 2
Quantity (thousands of
switchgears per week) (a) Complier Price and cost
(thousands of dollars per unit) Price and cost
(thousands of dollars per unit) FIGURE 13.6 $1,500 a unit and therefore an economic profit of
$4.5 million. This economic profit is the blue rectangle in part (b).
We’ve now described a second possible outcome
for the duopoly game: One of the firms cheats on the
collusive agreement. In this case, the industry output
is larger than the monopoly output and the industry
price is lower than the monopoly price. The total
economic profit made by the industry is also smaller
than the monopoly’s economic profit. Trick (the
cheat) makes an economic profit of $4.5 million, and
Gear (the complier) incurs an economic loss of $1
million. The industry makes an economic profit of
$3.5 million. This industry profit is $0.5 million less
than the economic profit that a monopoly would
make. But the profit is distributed unevenly. Trick
makes a bigger economic profit than it would under
the collusive agreement, while Gear incurs an economic loss.
A similar outcome would arise if Gear cheated and
Trick complied with the agreement. The industry
profit and price would be the same, but in this case,
Gear (the cheat) would make an economic profit of
$4.5 million and Trick (the complier) would incur an
economic loss of $1 million.
Let’s next see what happens if both firms cheat. ATC 10.0 7.5
profit 0 1 2
Quantity (thousands of
switchgears per week) (b) Cheat One firm, shown in part (a), complies with the agreement
and produces 2,000 units. The other firm, shown in part (b),
cheats on the agreement and increases its output to 3,000
units a week. Given the market demand curve, shown in
part (c), and with a total production of 5,000 units a week,
animation Price (thousands of dollars per unit) 296 10.0 7.5 D
output 0 1 2 3 4
Quantity (thousands of
switchgears per week) (c) Industry the price falls to $7,500 a unit. At this price, the complier in
part (a) incurs an economic loss of $1 million ($500 per
unit 2,000 units), shown by the red rectangle. In part (b),
the cheat makes an economic profit of $4.5 million ($1,500
per unit 3,000 units), shown by the blue rectangle. 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 297 O ligopoly Games Both Firms Cheat Suppose that both firms cheat and
that each firm behaves like the cheating firm that we
have just analyzed. Each tells the other that it is
unable to sell its output at the going price and that it
plans to cut its price. But because both firms cheat,
each will propose a successively lower price. As long as
price exceeds marginal cost, each firm has an incentive
to increase its production—to cheat. Only when price
equals marginal cost is there no further incentive to
cheat. This situation arises when the price has reached
$6,000. At this price, marginal cost equals price. Also,
price equals minimum average total cost. At a price
less than $6,000, each firm incurs an economic loss.
At a price of $6,000, each firm covers all its costs and
makes zero economic profit. Also, at a price of
$6,000, each firm wants to produce 3,000 units a
week, so the industry output is 6,000 units a week.
Given the demand conditions, 6,000 units can be sold
at a price of $6,000 each.
Figure 13.7 illustrates the situation just described.
Each firm, in part (a), produces 3,000 units a week,
and its average total cost is a minimum ($6,000 per
unit). The market as a whole, in part (b), operates at
the point at which the market demand curve (D)
intersects the industry marginal cost curve (MCI).
Each firm has lowered its price and increased its output to try to gain an advantage over the other firm.
Each has pushed this process as far as it can without
incurring an economic loss.
We have now described a third possible outcome
of this duopoly game: Both firms cheat. If both firms 10 MC ATC 6 0 1 cheat on the collusive agreement, the output of each
firm is 3,000 units a week and the price is $6,000 a
unit. Each firm makes zero economic profit.
The Payoff Matrix Now that we have described the
strategies and payoffs in the duopoly game, we can
summarize the strategies and the payoffs in the form
of the game’s payoff matrix. Then we can find the
Table 13.2 sets out the payoff matrix for this
game. It is constructed in the same way as the payoff
matrix for the prisoners’ dilemma in Table 13.1. The
squares show the payoffs for the two firms—Gear
and Trick. In this case, the payoffs are profits. (For
the prisoners’ dilemma, the payoffs were losses.)
The table shows that if both firms cheat (top left),
they achieve the perfectly competitive outcome—
each firm makes zero economic profit. If both firms
comply (bottom right), the industry makes the
monopoly profit and each firm makes an economic
profit of $2 million. The top right and bottom left
squares show the payoff if one firm cheats while the
other complies. The firm that cheats makes an economic profit of $4.5 million, and the one that complies incurs a loss of $1 million.
Nash Equilibrium in the Duopolists’ Dilemma The
duopolists have a dilemma like the prisoners’
dilemma. Do they comply or cheat? To answer this
question, we must find the Nash equilibrium. Both Firms Cheat 2
Quantity (thousands of
switchgears per week) (a) Individual firm animation Price and cost
(thousands of dollars per unit) Price and cost
(thousands of dollars per unit) FIGURE 13.7 297 10 MC I 6
Both firms cheating
outcome 0 (b) Industry 1 2 D 3
Quantity (thousands of
switchgears per week) If both firms cheat by increasing production,
the collusive agreement collapses. The limit
to the collapse is the competitive equilibrium. Neither firm will cut its price below
$6,000 (minimum average total cost)
because to do so will result in losses. In part
(a), each firm produces 3,000 units a week
at an average total cost of $6,000. In part
(b), with a total production of 6,000 units,
the price falls to $6,000. Each firm now
makes zero economic profit. This output
and price are the ones that would prevail in
a competitive industry. 9160335_CH13_p287-314.qxd 298 6/22/09 9:05 AM Page 298 CHAPTER 13 Oligopoly TABLE 13.2 Duopoly Payoff Matrix Other Oligopoly Games Gear's strategies
Cheat Comply $0 –$1.0m Cheat
$0 +$4.5m Trick's
+$4.5m +$2m Firms in oligopoly must decide whether to mount
expensive advertising campaigns; whether to modify
their product; whether to make their product more
reliable and more durable; whether to price discriminate and, if so, among which groups of customers
and to what degree; whether to undertake a large
research and development (R&D) effort aimed at
lowering production costs; and whether to enter or
leave an industry.
All of these choices can be analyzed as games that
are similar to the one that we’ve just studied. Let’s
look at one example: an R&D game. Comply
–$1.0m +$2m An R&D Game
Procter & Gamble Versus Kimberly-Clark Each square shows the payoffs from a pair of actions.
For example, if both firms comply with the collusive
agreement, the payoffs are recorded in the bottom
right square. The red triangle shows Gear’s payoff,
and the blue triangle shows Trick’s. In Nash equilibrium, both firms cheat. Look at things from Gear’s point of view. Gear
reasons as follows: Suppose that Trick cheats. If I
comply, I will incur an economic loss of $1 million.
If I also cheat, I will make zero economic profit. Zero
is better than minus $1 million, so I’m better off if I
cheat. Now suppose Trick complies. If I cheat, I will
make an economic profit of $4.5 million, and if I
comply, I will make an economic profit of $2 million. A $4.5 million profit is better than a $2 million
profit, so I’m better off if I cheat. So regardless of
whether Trick cheats or complies, it pays Gear to
cheat. Cheating is Gear’s best strategy.
Trick comes to the same conclusion as Gear
because the two firms face an identical situation. So
both firms cheat. The Nash equilibrium of the duopoly game is that both firms cheat. And although the
industry has only two firms, they charge the same
price and produce the same quantity as those in a
competitive industry. Also, as in perfect competition,
each firm makes zero economic profit.
This conclusion is not general and will not always
arise. We’ll see why not by looking first at some other
games that are like the prisoners’ dilemma. Then we’ll
broaden the types of games we consider. Disposable diapers have been around for a bit more
than 40 years. Procter & Gamble (which has a
40 percent market share with Pampers) and KimberlyClark (which has a 33 percent market share with
Huggies) have always been the market leaders.
When the disposable diaper was first introduced,
it had to be cost-effective in competition with
reusable, laundered diapers. A costly research and
development effort resulted in the development of
machines that could make disposable diapers at a low
enough cost to achieve that initial competitive edge.
But new firms tried to get into the business and take
market share away from the two industry leaders, and
the industry leaders themselves battled each other to
maintain or increase their own market shares.
During the early 1990s, Kimberly-Clark was the first
to introduce Velcro closures. And in 1996, Procter &
Gamble was the first to introduce “breathable” diapers.
The key to success in this industry (as in any other)
is to design a product that people value highly relative
to the cost of producing it. The firm that creates the
most highly valued product and also develops the
least-cost technology for producing it gains a competitive edge, undercutting the rest of the market, increasing its market share, and increasing its profit.
But the R&D that must be undertaken to improve
product quality and cut cost is itself costly. So the
cost of R&D must be deducted from the profit
resulting from the increased market share that lower
costs achieve. If no firm does R&D, every firm can
be better off, but if one firm initiates the R&D activity, all must follow. 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 299 O ligopoly Games Table 13.3 illustrates the dilemma (with hypothetical
numbers) for the R&D game that Kimberly-Clark and
Procter & Gamble play. Each firm has two strategies:
Spend $25 million a year on R&D or spend nothing on
R&D. If neither firm spends on R&D, they make a
joint profit of $100 million: $30 million for KimberlyClark and $70 million for Procter & Gamble (bottom
right of the payoff matrix). If each firm conducts R&D,
market shares are maintained but each firm’s profit is
lower by the amount spent on R&D (top left square of
the payoff matrix). If Kimberly-Clark pays for R&D
but Procter & Gamble does not, Kimberly-Clark gains
a large part of Procter & Gamble’s market. KimberlyClark profits, and Procter & Gamble loses (top right
square of the payoff matrix). Finally, if Procter &
Gamble conducts R&D and Kimberly-Clark does not,
Procter & Gamble gains market share from KimberlyClark, increasing its profit, while Kimberly-Clark incurs
a loss (bottom left square).
TABLE 13.3 Pampers Versus Huggies:
An R&D Game
Procter & Gamble's strategies
R&D No R&D $45m –$10m R&D
$5m +$85m KimberlyClark's
+$85m +$70m No R&D
–$10m +$30m If both firms undertake R&D, their payoffs are those
shown in the top left square. If neither firm undertakes
R&D, their payoffs are in the bottom right square. When
one firm undertakes R&D and the other one does not,
their payoffs are in the top right and bottom left squares.
The red triangle shows Procter & Gamble’s payoff, and
the blue triangle shows Kimberly-Clark’s. The Nash equilibrium for this game is for both firms to undertake R&D.
The structure of this game is the same as that of the prisoners’ dilemma. 299 Confronted with the payoff matrix in Table 13.3,
the two firms calculate their best strategies. KimberlyClark reasons as follows: If Procter & Gamble does
not undertake R&D, we will make $85 million if we
do and $30 million if we do not; so it pays us to conduct R&D. If Procter & Gamble conducts R&D, we
will lose $10 million if we don’t and make $5 million
if we do. Again, R&D pays off. So conducting R&D
is the best strategy for Kimberly-Clark. It pays,
regardless of Procter & Gamble’s decision.
Procter & Gamble reasons similarly: If KimberlyClark does not undertake R&D, we will make $70
million if we follow suit and $85 million if we conduct R&D. It therefore pays to conduct R&D. If
Kimberly-Clark does undertake R&D, we will make
$45 million by doing the same and lose $10 million
by not doing R&D. Again, it pays us to conduct R&D.
So for Procter & Gamble, R&D is also the best strategy.
Because R&D is the best strategy for both players,
it is the Nash equilibrium. The outcome of this game
is that both firms conduct R&D. They make less
profit than they would if they could collude to
achieve the cooperative outcome of no R&D.
The real-world situation has more players than
Kimberly-Clark and Procter & Gamble. A large
number of other firms share a small portion of the
market, all of them ready to eat into the market share
of Procter & Gamble and Kimberly-Clark. So the
R&D efforts by these two firms not only serve the
purpose of maintaining shares in their own battle but
also help to keep barriers to entry high enough to
preserve their joint market share. The Disappearing Invisible Hand
All the games that we’ve studied are versions of the
prisoners’ dilemma. The essence of that game lies in
the structure of its payoffs. The worst possible outcome for each player arises from cooperating when
the other player cheats. The best possible outcome, for
each player to cooperate, is not a Nash equilibrium
because it is in neither player’s self-interest to cooperate
if the other one cooperates. It is this failure to achieve
the best outcome for both players—the best social
outcome if the two players are the entire economy—
that led John Nash to claim (as he was portrayed as
doing in the movie A Beautiful Mind ) that he had
challenged Adam Smith’s idea that we are always
guided, as if by an invisible hand, to promote the
social interest when we are pursuing our self-interest. 9160335_CH13_p287-314.qxd 300 6/22/09 9:05 AM Page 300 CHAPTER 13 Oligopoly A Game of Chicken
The Nash equilibrium for the prisoners’ dilemma is
called a dominant strategy equilibrium, which is an
equilibrium in which the best strategy of each player
is to cheat (confess) regardless of the strategy of the
other player. Not all games have such an equilibrium, and one that doesn’t is a game called
In a graphic, if disturbing, version of this game,
two cars race toward each other. The first driver to
swerve and avoid a crash is “chicken.” The payoffs are
a big loss for both if no one “chickens,” zero for the
chicken, and a gain for the player who stays the
course. If player 1 chickens, player 2’s best strategy is
to stay the course. And if player 1 stays the course,
player 2’s best strategy is to chicken.
For an economic form of this game, suppose the
R&D that creates a new diaper technology results in
information that cannot be kept secret or patented, so
both firms benefit from the R&D of either firm. The
chicken in this case is the firm that does the R&D.
Table 13.4 illustrates a payoff matrix for an R&D
game of chicken between Kimberly-Clark and Procter
& Gamble. Each firm has two strategies: Do the
R&D (and “chicken”) or do not do the R&D (and
If neither “chickens,” there is no R&D and each
firm makes zero additional profit. If each firm conducts R&D—each “chickens”—each firm makes $5
million (the profit from the new technology minus
the cost of the research). If one of the firms does the
R&D, the payoffs are $1 million for the chicken and
$10 million for the one who stands firm.
Confronted with the payoff matrix in Table 13.4,
the two firms calculate their best strategies. KimberlyClark is better off doing R&D if Procter & Gamble
does not undertake it. Procter & Gamble is better off
doing R&D if Kimberly-Clark doesn’t do it. There are
two equilibrium outcomes: One firm does the R&D,
but we can’t predict which firm it will be.
You can see that it isn’t a Nash equilibrium if no
firm does the R&D because one firm would then be
better off doing it. And you can see that it isn’t a
Nash equilibrium if both firms do the R&D because
then one firm would be better off not doing it.
The firms could toss a coin or use some other random device to make a decision in this game. In some
circumstances, such a strategy—called a mixed strategy—is actually better for both firms than choosing
any of the strategies we’ve considered. TABLE 13.4 An R&D Game of Chicken
Procter & Gamble's strategies
R&D No R&D $5m $10m R&D
$5m $1m KimberlyClark's
$1m $0 No R&D
$10m $0 If both firms undertake R&D, their payoffs are those shown in
the top left square. If neither firm undertakes R&D, their payoffs are in the bottom right square. When one firm undertakes R&D and the other one does not, their payoffs are in
the top right and bottom left squares. The red triangle shows
Procter & Gamble’s payoff, and the blue triangle shows
Kimberly-Clark’s. The equilibrium for this R&D game of
chicken is for only one firm to undertake R&D. We cannot
tell which firm will do the R&D and which will not. Review Quiz ◆
2 3 4
5 6 What are the common features of all games?
Describe the prisoners’ dilemma game and
explain why the Nash equilibrium delivers a
bad outcome for both players.
Why does a collusive agreement to restrict output and raise price create a game like the prisoners’ dilemma?
What creates an incentive for firms in a collusive
agreement to cheat and increase production?
What is the equilibrium strategy for each firm in
a duopolists’ dilemma and why do the firms not
succeed in colluding to raise the price and profits?
Describe two structures of payoffs for an R&D
game and contrast the prisoners’ dilemma and
the chicken game.
Work Study Plan 13.3
and get instant feedback. 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 301 Repeated Games and Sequential Games ◆ Repeated Games
and Sequential Games
The games that we’ve studied are played just once.
In contrast, many real-world games are played
repeatedly. This feature of games turns out to enable
real-world duopolists to cooperate, collude, and
make a monopoly profit.
Another feature of the games that we’ve studied is
that the players move simultaneously. But in many
real-world situations, one player moves first and then
the other moves—the play is sequential rather than
simultaneous. This feature of real-world games creates a large number of possible outcomes.
We’re now going to examine these two aspects of
strategic decision-making. A Repeated Duopoly Game
If two firms play a game repeatedly, one firm has
the opportunity to penalize the other for previous
“bad” behavior. If Gear cheats this week, perhaps
Trick will cheat next week. Before Gear cheats this
week, won’t it consider the possibility that Trick will
cheat next week? What is the equilibrium of this
Actually, there is more than one possibility. One is
the Nash equilibrium that we have just analyzed. Both
players cheat, and each makes zero economic profit
forever. In such a situation, it will never pay one of
the players to start complying unilaterally because to
do so would result in a loss for that player and a profit
for the other. But a cooperative equilibrium in which
the players make and share the monopoly profit is
A cooperative equilibrium might occur if cheating
is punished. There are two extremes of punishment.
The smallest penalty is called “tit for tat.” A tit-for-tat
strategy is one in which a player cooperates in the current period if the other player cooperated in the previous period, but cheats in the current period if the
other player cheated in the previous period. The most
severe form of punishment is called a trigger strategy.
A trigger strategy is one in which a player cooperates if
the other player cooperates but plays the Nash equilibrium strategy forever thereafter if the other player
In the duopoly game between Gear and Trick, a
tit-for-tat strategy keeps both players cooperating and
making monopoly profits. Let’s see why with an example. 301 Table 13.5 shows the economic profit that Trick
and Gear will make over a number of periods under
two alternative sequences of events: colluding and
cheating with a tit-for-tat response by the other firm.
If both firms stick to the collusive agreement in
period 1, each makes an economic profit of $2 million. Suppose that Trick contemplates cheating in
period 1. The cheating produces a quick $4.5 million
economic profit and inflicts a $1 million economic
loss on Gear. But a cheat in period 1 produces a
response from Gear in period 2. If Trick wants to get
back into a profit-making situation, it must return to
the agreement in period 2 even though it knows that
Gear will punish it for cheating in period 1. So in
period 2, Gear punishes Trick and Trick cooperates.
Gear now makes an economic profit of $4.5 million,
and Trick incurs an economic loss of $1 million.
Adding up the profits over two periods of play, Trick
would have made more profit by cooperating—
$4 million compared with $3.5 million.
What is true for Trick is also true for Gear. Because
each firm makes a larger profit by sticking with the
collusive agreement, both firms do so and the
monopoly price, quantity, and profit prevail.
In reality, whether a cartel works like a one-play
game or a repeated game depends primarily on the
TABLE 13.5 Cheating with Punishment
tit-for-tat C ollude
of play Trick’s
profit (millions of dollars) Trick’s
profit (millions of dollars) 1 2 2 4.5 –1.0 2 2 2 –1.0 4.5 3 2 2 2.0 2.0 4 . . . . If duopolists repeatedly collude, each makes a profit of
$2 million per period of play. If one player cheats in
period 1, the other player plays a tit-for-tat strategy and
cheats in period 2. The profit from cheating can be
made for only one period and must be paid for in the
next period by incurring a loss. Over two periods of
play, the best that a duopolist can achieve by cheating is
a profit of $3.5 million, compared to an economic profit
of $4 million by colluding. 9160335_CH13_p287-314.qxd 302 6/22/09 9:05 AM Page 302 CHAPTER 13 Oligopoly number of players and the ease of detecting and punishing cheating. The larger the number of players, the
harder it is to maintain a cartel.
Games and Price Wars A repeated duopoly game
can help us understand real-world behavior and, in
particular, price wars. Some price wars can be interpreted as the implementation of a tit-for-tat strategy.
But the game is a bit more complicated than the one
we’ve looked at because the players are uncertain
about the demand for the product.
Playing a tit-for-tat strategy, firms have an incentive to stick to the monopoly price. But fluctuations
in demand lead to fluctuations in the monopoly price,
and sometimes, when the price changes, it might seem
to one of the firms that the price has fallen because
the other has cheated. In this case, a price war will
break out. The price war will end only when each firm
is satisfied that the other is ready to cooperate again.
There will be cycles of price wars and the restoration
of collusive agreements. Fluctuations in the world
price of oil might be interpreted in this way.
Some price wars arise from the entry of a small
number of firms into an industry that had previously been a monopoly. Although the industry has a
small number of firms, the firms are in a prisoners’
dilemma and they cannot impose effective penalties
for price cutting. The behavior of prices and outputs
in the computer chip industry during 1995 and
1996 can be explained in this way. Until 1995, the
market for Pentium chips for IBM-compatible computers was dominated by one firm, Intel Corporation, which was able to make maximum economic
profit by producing the quantity of chips at which
marginal cost equaled marginal revenue. The price
of Intel’s chips was set to ensure that the quantity
demanded equaled the quantity produced. Then in
1995 and 1996, with the entry of a small number of
new firms, the industry became an oligopoly. If the
firms had maintained Intel’s price and shared the
market, together they could have made economic
profits equal to Intel’s profit. But the firms were in a
prisoners’ dilemma, so prices fell toward the competitive level.
Let’s now study a sequential game. There are many
such games, and the one we’ll examine is among the
simplest. It has an interesting implication and it will
give you the flavor of this type of game. The sequential game that we’ll study is an entry game in a contestable market. A Sequential Entry Game
in a Contestable Market
If two firms play a sequential game, one firm makes
a decision at the first stage of the game and the
other makes a decision at the second stage.
We’re going to study a sequential game in a
contestable market—a market in which firms can enter
and leave so easily that firms in the market face competition from potential entrants. Examples of contestable markets are routes served by airlines and by
barge companies that operate on the major waterways. These markets are contestable because firms
could enter if an opportunity for economic profit
arose and could exit with no penalty if the opportunity for economic profit disappeared.
If the Herfindahl-Hirschman Index (p. 215) is
used to determine the degree of competition, a contestable market appears to be uncompetitive. But a
contestable market can behave as if it were perfectly
competitive. To see why, let’s look at an entry game
for a contestable air route.
A Contestable Air Route Agile Air is the only firm
operating on a particular route. Demand and cost
conditions are such that there is room for only one
airline to operate. Wanabe Inc. is another airline that
could offer services on the route.
We describe the structure of a sequential game by
using a game tree like that in Fig. 13.8. At the first
stage, Agile Air must set a price. Once the price is set
and advertised, Agile can’t change it. That is, once set,
Agile’s price is fixed and Agile can’t react to Wanabe’s
entry decision. Agile can set its price at either the
monopoly level or the competitive level.
At the second stage, Wanabe must decide
whether to enter or to stay out. Customers have no
loyalty (there are no frequent-flyer programs) and
they buy from the lowest-price firm. So if Wanabe
enters, it sets a price just below Agile’s and takes all
Figure 13.8 shows the payoffs from the various
decisions (Agile’s in the red triangles and Wanabe’s in
the blue triangles).
To decide on its price, Agile’s CEO reasons as follows: Suppose that Agile sets the monopoly price. If
Wanabe enters, it earns 90 (think of all payoff numbers as thousands of dollars). If Wanabe stays out, it
earns nothing. So Wanabe will enter. In this case
Agile will lose 50. 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 303 R epeated Games and Sequential Games FIGURE 13.8 303 Agile Versus Wanabe: A Sequential Entry Game in a Contestable Market First stage Second stage Payoffs
Monopoly price –50
Stay out Agile 100
–10 Enter –50 Wanabe 0 Competitive price
Stay out If Agile sets the monopoly price, Wanabe makes 90 (thousand dollars) by entering and earns nothing by staying out.
So if Agile sets the monopoly price, Wanabe enters. 0 If Agile sets the competitive price, Wanabe earns nothing if it stays out and incurs a loss if it enters. So if Agile
sets the competitive price, Wanabe stays out. animation Now suppose that Agile sets the competitive price.
If Wanabe stays out, it earns nothing, and if it enters,
it loses 10, so Wanabe will stay out. In this case, Agile
will make zero economic profit.
Agile’s best strategy is to set its price at the competitive level and make zero economic profit. The
option of earning 100 by setting the monopoly price
with Wanabe staying out is not available to Agile. If
Agile sets the monopoly price, Wanabe enters, undercuts Agile, and takes all the business.
In this example, Agile sets its price at the competitive level and makes zero economic profit. A less
costly strategy, called limit pricing, sets the price at the
highest level that inflicts a loss on the entrant. Any
loss is big enough to deter entry, so it is not always
necessary to set the price as low as the competitive
price. In the example of Agile and Wanabe, at the
competitive price, Wanabe incurs a loss of 10 if it
enters. A smaller loss would still keep Wanabe out.
This game is interesting because it points to the possibility of a monopoly behaving like a competitive
industry and serving the social interest without regulation. But the result is not general and depends on one
crucial feature of the setup of the game: At the second
stage, Agile is locked in to the price set at the first stage.
If Agile could change its price in the second stage,
it would want to set the monopoly price if Wanabe
stayed out—100 with the monopoly price beats zero
with the competitive price. But Wanabe can figure out what Agile would do, so the price set at the first stage
has no effect on Wanabe. Agile sets the monopoly
price and Wanabe might either stay out or enter.
We’ve looked at two of the many possible repeated
and sequential games, and you’ve seen how these
types of games can provide insights into the complex
forces that determine prices and profits. Review Quiz ◆
1 2 If a prisoners’ dilemma game is played repeatedly, what punishment strategies might the
players employ and how does playing the game
repeatedly change the equilibrium?
If a market is contestable, how does the equilibrium differ from that of a monopoly?
Work Study Plan 13.4
and get instant feedback. So far, we’ve studied oligopoly with unregulated
market power. Firms like Trick and Gear are free to
collude to maximize their profit with no concern for
the consumer or the law.
But when firms collude to achieve the monopoly
outcome, they also have the same effects on efficiency
and the social interest as monopoly. Profit is made at
the expense of consumer surplus and a deadweight
loss arises. Your next task is to see how U.S. antitrust
law limits market power. 9160335_CH13_p287-314.qxd 304 6/22/09 9:05 AM Page 304 CHAPTER 13 Oligopoly ◆ Antitrust Law
is the law that regulates oligopolies and
prevents them from becoming monopolies or
behaving like monopolies. Two government agencies
cooperate to enforce the antitrust laws: the Federal
Trade Commission and the Antitrust Division of
the U.S. Department of Justice.
Antitrust law The Antitrust Laws
The two main antitrust laws are
■ The Sherman Act, 1890
■ The Clayton Act, 1914.
The Sherman Act The Sherman Act made it a felony to create or attempt to create a monopoly or a cartel.
During the 1880s, lawmakers and the general
public were outraged and disgusted by the practices
of some of the big-name leaders of American business. The actions of J.P. Morgan, John D.
Rockefeller, and W.H. Vanderbilt led to them being
called the “robber barons.” It turns out that the most
lurid stories of the actions of these great American
capitalists were not of their creation of monopoly
power to exploit consumers but of their actions to
damage each other.
Nevertheless, monopolies that damaged the consumer interest did emerge. For example, John D.
Rockefeller had a virtual monopoly in the market for
Table 13.6 summarizes the two main provisions of
the Sherman Act. Section 1 of the act is precise: TABLE 13.6 The Sherman Act of 1890 Section 1:
Every contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce
among the several States, or with foreign nations, is hereby declared to be illegal. Section 2:
Every person who shall monopolize, or attempt to monopolize, or combine or conspire with any other person or
persons, to monopolize any part of the trade or commerce among the several States, or with foreign nations,
shall be deemed guilty of a felony. Conspiring with others to restrict competition is illegal. But Section 2 is general and imprecise. Just what
is an “attempt to monopolize”?
The Clayton Act The Clayton Act, which was passed in response to a wave of mergers that
occurred at the beginning of the twentieth century,
provided the answer to the question left dangling
by the Sherman Act: It defined the “attempt to
monopolize.” The Clayton Act supplemented the
Sherman Act and strengthened and clarified the
When Congress passed the Clayton Act, it also
established the Federal Trade Commission, the federal
agency charged with the task of preventing monopoly
practices that damage the consumer interest.
Two amendments to the Clayton Act, the
Robinson-Patman Act of 1936 and the CellerKefauver Act of 1950, outlaw specific practices and
provided even greater precision to the antitrust law.
Table 13.7 describes these practices and summarizes
the main provisions of these three acts. TABLE 13.7 The Clayton Act and
Its Amendments Clayton Act 1914 Robinson-Patman Act 1936 Celler-Kefauver Act 1950 These acts prohibit the following practices only if they
substantially lessen competition or create monopoly:
1. Price discrimination
2. Contracts that require other goods to be bought from
the same firm (called tying arrangements)
3. Contracts that require a firm to buy all its requirements
of a particular item from a single firm (called requirements contracts)
4. Contracts that prevent a firm from selling competing
items (called exclusive dealing)
5. Contracts that prevent a buyer from reselling a product
outside a specified area (called territorial confinement )
6. Acquiring a competitor’s shares or assets
7. Becoming a director of a competing firm 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 305 Antitrust Law Price Fixing Always Illegal
Colluding to fix the price is always a violation of the
antitrust law. If the Justice Department can prove
the existence of a price fixing cartel, defendants can
offer no acceptable excuse.
The predictions of the effects of price fixing that
you saw in the previous sections of this chapter provide the reasons for the unqualified attitude toward
price fixing. A duopoly cartel can maximize profit
and behave like a monopoly. To achieve the monopoly outcome, the cartel restricts production and fixes
the price at the monopoly level. The consumer suffers
because consumer surplus shrinks. And the outcome
is inefficient because a deadweight loss arises.
It is for these reasons that the law declares that
all price fixing is illegal. No excuse can justify the
Other antitrust practices are more controversial
and generate debate among lawyers and economists.
We’ll examine three of these practices. Three Antitrust Policy Debates
The three practices that we’ll examine are
■ Resale price maintenance
Predatory pricing 305 Resale price maintenance is inefficient if it enables dealers to charge the
monopoly price. By setting and enforcing the resale
price, the manufacturer might be able to achieve the
Inefficient Resale Price Maintenance Efficient Resale Price Maintenance Resale price maintenance might be efficient if it enables a manufacturer to
induce dealers to provide the efficient standard of
service. Suppose that SilkySkin wants shops to demonstrate the use of its new unbelievable moisturizing
cream in an inviting space. With resale price maintenance, SilkySkin can offer all the retailers the same
incentive and compensation. Without resale price
maintenance, a cut-price drug store might offer
SilkySkin products at a low price. Buyers would then
have an incentive to visit a high-price shop and get the
product demonstrated and then buy from the low-price
shop. The low-price shop would be a free rider (like the
consumer of a public good in Chapter 15), and an
inefficient level of service would be provided.
SilkySkin could pay a fee to retailers that provide
good service and leave the resale price to be determined by the competitive forces of supply and
demand. But it might be too costly for SilkySkin to
monitor shops and ensure that they provided the
desired level of service. Tying Arrangements A tying arrangement is an agreeResale Price Maintenance Most manufacturers sell their products to the final consumer indirectly
through a wholesale and retail distribution system.
Resale price maintenance occurs when a manufacturer
agrees with a distributor on the price at which the
product will be resold.
Resale price maintenance (also called vertical price
fixing) agreements are illegal under the Sherman Act.
But it isn’t illegal for a manufacturer to refuse to supply a retailer who doesn’t accept guidance on what
the price should be.
Nor is it illegal to set a minimum retail price provided it is not anticompetitive. In 2007, the Supreme
Court ruled that a handbag manufacturer could
impose a minimum retail price on a Dallas store,
Kay's Kloset. Since that ruling, many manufacturers
have imposed minimum retail prices. The practice is
judged on a case-by-case basis.
Does resale price maintenance create an inefficient
or efficient use of resources? Economists can be
found on both sides of this question. ment to sell one product only if the buyer agrees to
buy another, different product. With tying, the only
way the buyer can get the one product is to also buy
the other product. Microsoft has been accused of
tying Internet Explorer and Windows. Textbook publishers sometimes tie a Web site and a textbook and
force students to buy both. (You can’t buy the book
you’re now reading, new, without the Web site. But
you can buy the Web site access without the book, so
these products are not tied.)
Could textbook publishers make more money by
tying a book and access to a Web site? The answer
is sometimes but not always. Suppose that you and
other students are willing to pay $80 for a book
and $20 for access to a Web site. The publisher can
sell these items separately for these prices or bundled for $100. The publisher does not gain from
But now suppose that you and only half of the students are willing to pay $80 for a book and $20 for a
Web site and the other half of the students are willing 9160335_CH13_p287-314.qxd 306 6/22/09 9:05 AM Page 306 CHAPTER 13 Oligopoly to pay $80 for a Web site and $20 for a book. Now if
the two items are sold separately, the publisher can
charge $80 for the book and $80 for the Web site.
Half the students buy the book but not the Web site,
and the other half buy the Web site but not the book.
But if the book and Web site are bundled for $100,
everyone buys the bundle and the publisher makes an
extra $20 per student. In this case, bundling has
enabled the publisher to price discriminate.
There is no simple, clear-cut test of whether a firm
is engaging in tying or whether, by doing so, it has
increased its market power and profit and created
Predatory Pricing Predatory pricing is setting a low price to drive competitors out of business with the
intention of setting a monopoly price when the com- An Antitrust Showcase
The United States Versus Microsoft
In 1998, the Antitrust Division of the U.S.
Department of Justice along with the Departments
of Justice of a number of states charged Microsoft,
the world’s largest producer of software for personal
computers, with violations of both sections of the
A 78-day trial followed that pitched two prominent MIT economics professors against each other,
Franklin Fisher for the government and Richard
Schmalensee for Microsoft.
The Case Against Microsoft The claims against Microsoft were that it
Possessed monopoly power
■ Used predatory pricing and tying arrangements
■ Used other anticompetitive practices
It was claimed that with 80 percent of the market
for PC operating systems, Microsoft had excessive
monopoly power. This monopoly power arose from
two barriers to entry: economies of scale and network
economies. Microsoft’s average total cost falls as production increases (economies of scale) because the
fixed cost of developing an operating system such as
Windows is large while the marginal cost of producing one copy of Windows is small. Further, as the
number of Windows users increases, the range of
Windows applications expands (network economies),
so a potential competitor would need to produce not
■ petition has gone. John D. Rockefeller’s Standard Oil
Company was the first to be accused of this practice in
the 1890s, and it has been claimed often in antitrust
cases since then. Predatory pricing is an attempt to create a monopoly and as such it is illegal under Section 2
of the Sherman Act.
It is easy to see that predatory pricing is an idea, not
a reality. Economists are skeptical that predatory pricing occurs. They point out that a firm that cuts its
price below the profit-maximizing level loses during
the low-price period. Even if it succeeds in driving its
competitors out of business, new competitors will
enter as soon as the price is increased, so any potential
gain from a monopoly position is temporary. A high
and certain loss is a poor exchange for a temporary and
uncertain gain. No case of predatory pricing has been
only a competing operating system but also an entire
range of supporting applications as well.
When Microsoft entered the Web browser market
with its Internet Explorer, it offered the browser for a
zero price. This price was viewed as predatory pricing.
Microsoft integrated Internet Explorer with Windows
so that anyone who uses this operating system would
not need a separate browser such as Netscape
Navigator. Microsoft’s competitors claimed that this
practice was an illegal tying arrangement.
Microsoft’s Response Microsoft challenged all these
claims. It said that Windows was vulnerable to competition from other operating systems such as Linux
and Apple’s Mac OS and that there was a permanent
threat of competition from new entrants.
Microsoft claimed that integrating Internet
Explorer with Windows provided a single, unified
product of greater consumer value like a refrigerator
with a chilled water dispenser or an automobile with
a CD player.
The Outcome The court agreed that Microsoft was in violation of the Sherman Act and ordered that it be
broken into two firms: an operating systems producer
and an applications producer. Microsoft successfully
appealed this order. In the final judgment, though,
Microsoft was ordered to disclose to other software
developers details of how its operating system works,
so that they could compete effectively against
Microsoft. In the summer of 2002, Microsoft began
to comply with this order. 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 307 Antitrust Law Mergers and Acquisitions and a merger in this market that would increase the
index by 100 points is challenged by the FTC. An
index above 1,800 indicates a concentrated market,
and a merger in this market that would increase the
index by 50 points is challenged. You can see an
application of these guidelines in the box below. Mergers, which occur when two or more firms agree
to combine to create one larger firm, and acquisitions, which occur when one firm buys another
firm, are common events. Mergers occurred when
Chrysler and the German auto producer DaimlerBenz combined to form DaimlerChrysler and when
the Belgian beer producer InBev bought the U.S.
brewing giant Anheuser-Busch and created a new
combined company, Anheuser-Busch InBev. An
acquisition occurred when Rupert Murdoch’s News
Corp bought Myspace.
The mergers and acquisitions that occur don’t create a monopoly. But two (or more) firms might be
tempted to try to merge so that they can gain market
power and operate like a monopoly. If such a situation arises, the Federal Trade Commission (FTC)
takes an interest in the move and stands ready to
block the merger.
To determine which mergers it will examine and
possibly block, the FTC uses guidelines, one of
which is the Herfindahl-Hirschman Index (HHI)
(see Chapter 10, pp. 238–239).
A market in which the HHI is less than 1,000 is
regarded as competitive. An index between 1,000 and
1,800 indicates a moderately concentrated market, Merger Guidelines Review Quiz ◆
5 What are the two main antitrust laws and when
were they enacted?
When is price fixing not a violation of the
What is an attempt to monopolize an industry?
What are resale price maintenance, tying
arrangements, and predatory pricing?
Under what circumstances is a merger unlikely
to be approved?
Work Study Plan 13.5
and get instant feedback. ◆ Oligopoly is a market structure that you often encounter in your daily life. Reading Between the
Lines on pp. 308–309 looks at a game played by Dell
and HP in the market for personal computers. Competitive Moderately Concentrated
concentrated FTC Takes the Fizz out of Soda Mergers
The FTC used its HHI guidelines to block proposed mergers in the market for soft drinks.
PepsiCo wanted to buy 7-Up and Coca-Cola
wanted to buy Dr Pepper. The market for carbonated soft drinks is highly concentrated. Coca-Cola
had a 39 percent share, PepsiCo had 28 percent, Dr
Pepper was next with 7 percent, followed by 7-Up
with 6 percent. One other producer, RJR, had a 5
percent market share. So the five largest firms in this
market had an 85 percent market share.
The PepsiCo and 7-Up merger would have increased
the HHI by more than 300 points. The Coca-Cola and
Dr Pepper merger would have increased it by more
than 500 points, and both mergers together would have
increased the index by almost 800 points.
The FTC decided that increases in the HHI of
these magnitudes were not in the social interest and
blocked the mergers. The figure summarizes the HHI
guideline and HHIs in the soft drinks market. 307 Challenge merger if index rises by more than: 100 points 0 1,000 50 points 1,800 3,000 4,000 3,000 4,000 Herfindahl-Hirschman Index (HHI)
Figure 1 The Merger Guidelines No mergers
Pepsi / 7-Up
Coke / Dr Pepper
Both mergers 0 1,000 2,000 Herfindahl-Hirschman Index (HHI)
Figure 2 Product Mergers in Soft Drinks 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 308 READING BETWEEN THE LINES Dell and HP in a Market Share Game
The Old Price-War Tactic May Not Faze
May 13, 2006 Dell is sharply reducing prices on its computers.
The tactic is classic, straight out of the playbook that made the company the world’s largest
computer maker. As overall demand for personal computers slows, lower your prices. Profit
margins will take a temporary hit, but the move would hurt competitors worse as you take
market share and enjoy revenue growth for years to come.
Dell did it in 2000 and it worked beautifully. But after Dell rolled out the plan last month,
knocking as much as $700 off a $1,200 Inspiron and $500 off a $1,079 Dimension desktop,
many of the securities analysts who follow the company, based in Round Rock, Tex., said
that this time around it could be folly. ...
What changed? ... More than anything else, Dell’s competitors have changed. In particular,
Hewlett-Packard is no longer the bloated and slow-moving company it was six years ago. ...
The most telling evidence of the new landscape for PCs was seen in statistics on worldwide shipments. While the industry grew 12.9 percent in the first three months of the year, ... Dell’s shipments grew 10.2 percent. It was the first time since analysts began tracking Dell that its shipments
grew more slowly than the industry’s. Hewlett’s shipments, meanwhile, grew 22.2 percent. ...
Inside Hewlett, however, there is a feeling that it can beat Dell without resorting to price
wars. ... The company has started an ambitious marketing campaign to make that point with
ads that proclaim, “the computer is personal again.” ...
The campaign ... will feature celebrities and how they individualize their computers ... [HP] has
added technology like QuickPlay, which lets a user view a DVD or listen to a CD without waiting for the laptop’s operating system to boot up. The ads will say, “Don’t boot. Play.” ...
Copyright 2006 The New York Times Company. Reprinted with permission. Further reproduction prohibited. Essence of the Story
■ Dell cut its prices in 2000 and increased its market
share and revenue in the years that followed. ■ In April 2006, Dell slashed its prices. ■ Experts say the price cut will not work as well today. ■ Hewlett-Packard (HP) is much stronger than it was six
years ago. 308 ■ Total PC shipments increased by 12.9 percent in the
first quarter of 2006: Dell’s shipments increased by
10.2 percent, and HP’s increased by 22.2 percent. ♦ HP says that it can beat Dell without a price cut. Instead
it will launch a campaign to market PCs with new and
improved features, such as one that plays DVDs and
CDs without booting the operating system. 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 309 Economic Analysis
■ The global PC market has many firms, but two firms
dominate the market: Dell and Hewlett-Packard (HP). ■ Figure 1 shows the shares in the global PC market. You
can see that Dell and HP are the two biggest players
but almost 50 percent of the market is served by small
firms, each with less than 4 percent of the market. ■ Table 1 shows the payoff matrix (millions of dollars of
profit) for the game played by Dell and HP in 2000.
(The numbers are hypothetical.) ■ This game has a dominant strategy equilibrium similar
to that for the duopoly game on p. 353. ■ If HP cuts its price, Dell makes a larger profit by cutting
its price (+$20m versus –$10m), and if HP holds its
price constant, Dell again makes a larger profit by cutting its price (+$40m versus zero). ■ So Dell’s best strategy is to cut its price. ■ If Dell cuts its price, HP makes a larger profit by cutting
its price (+$5m versus –$20m), and if Dell holds its
price constant, HP again makes a larger profit by cutting its price (+$10m versus zero). ■ So Dell’s best strategy is to cut its price. ■ If Dell cuts its price, HP makes a larger profit by improving its marketing and design (+$20m versus
+$10m), and if Dell holds its price constant, HP again
makes a larger profit by improving its marketing and
design (+$40m versus +$20m). ■ So HP’s best strategy is to improve its marketing and
design. Dell's strategies
Cut price +$20m –$10m Cut price ■
■ Table 2 shows the payoffs from the game between Dell
and HP in 2006. ■ If HP cuts its price, Dell makes a larger profit by cutting
its price (+$10m versus –$10m), and if HP improves its
marketing and design, Dell again makes a larger profit
by cutting its price (+$5m versus –$20m). +$10m HP's
strategies This game, too, has a dominant strategy equilibrium. ■ +$5m So HP’s best strategy is to cut its price. +$40m $0 Hold price
–$20m $0 Table 1 The strategies and equilibrium in 2000 Dell's strategies
Cut price HP 16.3% Hold price Dell 16.1% Hold price +$10m –$10m Cut price
+$10m Lenovo 7.8% +$20m HP's
Toshiba 4.2% Others 49.6% Figure 1 Market shares in the PC market in 2006 +$5m Improve
+$20m –$20m +$40m Table 2 The strategies and equilibrium in 2006 309 9160335_CH13_p287-314.qxd 310 6/22/09 9:05 AM Page 310 CHAPTER 13 Oligopoly SUMMARY ◆ Key Points ■ What Is Oligopoly? (pp. 288–289)
■ Oligopoly is a market in which a small number of
firms compete. Antitrust Law (pp. 304–307)
■ Two Traditional Oligopoly Models (pp. 290–291)
■ ■ If rivals match price cuts but do not match price
hikes, each firm faces a kinked demand curve.
If one firm dominates a market, it acts like a
monopoly and the small firms act as price takers. ■ ■ Oligopoly Games (pp. 292–300)
■ ■ ■ ■ ■ Oligopoly is studied by using game theory, which
is a method of analyzing strategic behavior.
In a prisoners’ dilemma game, two prisoners acting
in their own self-interest harm their joint interest.
An oligopoly (duopoly) price-fixing game is a prisoners’ dilemma in which the firms might collude
In Nash equilibrium, both firms cheat and output
and price are the same as in perfect competition.
Firms’ decisions about advertising and R&D can
be studied by using game theory. In a sequential contestable market game, a small
number of firms can behave like firms in perfect
competition. ■ ■ ■ The first antitrust law, the Sherman Act, was
passed in 1890, and the law was strengthened in
1914 when the Clayton Act was passed and the
Federal Trade Commission was created.
All price-fixing agreements are violations of the
Sherman Act, and no acceptable excuse exists.
Resale price maintenance might be efficient if it
enables a producer to ensure the efficient level of
service by distributors.
Tying arrangements can enable a monopoly to
price discriminate and increase profit, but in many
cases, tying would not increase profit.
Predatory pricing is unlikely to occur because it
brings losses and only temporary potential gains.
The Federal Trade Commission uses guidelines
such as the Herfindahl-Hirschman Index to determine which mergers to investigate and possibly
block. Repeated Games and Sequential Games (pp. 301–303)
■ In a repeated game, a punishment strategy can
produce a cooperative equilibrium in which price
and output are the same as in a monopoly. Key Figures and Tables
Figure 13.5 Natural Oligopoly, 288
Costs and Demand, 294
Colluding to Make Monopoly
Profits, 295 Figure 13.6
Table 13.2 One Firm Cheats, 296
Both Firms Cheat, 297
Prisoners’ Dilemma Payoff Matrix, 293
Duopoly Payoff Matrix, 298 Key Terms
Antitrust law, 304
Collusive agreement, 294
Contestable market, 302
Cooperative equilibrium, 301
Dominant strategy equilibrium, 300 Duopoly, 288
Game theory, 292
Limit pricing, 303
Nash equilibrium, 293
Payoff matrix, 292 Predatory pricing, 306
Resale price maintenance, 305
Tying arrangement, 305 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 311 Problems and Applications PROBLEMS and APPLICATIONS 311 ◆ Work problems 1–11 in Chapter 13 Study Plan and get instant feedback.
Work problems 12–19 as Homework, a Quiz, or a Test if assigned by your instructor. 1. Two firms make most of the chips that power a
PC: Intel and Advanced Micro Devices. What
makes the market for PC chips a duopoly?
Sketch the market demand curve and cost curves
that describe the situation in this market and
that prevent other firms from entering.
2. The price at which Wal-Mart can buy flat-panel
TVs has fallen, and it is making a decision about
whether to cut its selling price. Wal-Mart believes
that if it cuts its price, all its competitors will cut
their prices, but if it raises its price, none of its
competitors will raise theirs.
a. Draw a figure to illustrate the situation that
Wal-Mart believes it faces in the market for
b. Do you predict that Wal-Mart will lower its
price of a flat-panel TV? Explain and illustrate
3. Big Joe’s Trucking has lower costs than the other
20 small truckers in the market. The market
operates like a dominant firm oligopoly and is
initially in equilibrium. Then the demand for
trucking services increases. Explain the effects of
the increase in demand on the price, output, and
economic profit of
a. Big Joe’s.
b. A typical small firm.
4. Consider a game with two players, who cannot
communicate, and in which each player is asked
a question. The players can answer the question
honestly or lie. If both answer honestly, each
receives $100. If one answers honestly and the
other lies, the liar receives $500 and the honest
player gets nothing. If both lie, then each receives
a. Describe the strategies and payoffs of this
b. Construct the payoff matrix.
c. What is the equilibrium of this game?
d. Compare this game to the prisoners’ dilemma.
Are the two games similar or different?
5. Soapy Inc. and Suddies Inc. are the only producers of soap powder. They collude and agree to
share the market equally. If neither firm cheats on the agreement, each makes $1 million profit.
If either firm cheats, the cheat makes a profit of
$1.5 million, while the complier incurs a loss of
$0.5 million. If both cheat, they break even.
Neither firm can monitor the other’s actions.
a. What are the strategies in this game?
b. Construct the payoff matrix for this game.
c. What is the equilibrium of this game if it is
played only once?
d. Is the equilibrium a dominant strategy equilibrium? Explain.
6. If Soapy Inc. and Suddies Inc. repeatedly play the
duopoly game that has the payoffs described in
problem 5 on each round of play,
a. What now are the strategies that each firm
b. Can the firms adopt a strategy that gives the
game a cooperative equilibrium?
c. Would one firm still be tempted to cheat in a
cooperative equilibrium? Explain your answer.
7. Oil City
In the late ’90s, Reliance spent $6 billion and
employed 75,000 workers to build a world-class
oil refinery at Jamnagar, India. … Now Reliance
is more than doubling the size of the facility,
which … will claim the title of the world’s
biggest … with an output of 1.2 million gallons
of gasoline per day, or about 5% of global capacity. … Reliance plans to aim Jamnagar’s spigots
westward, at the U.S. and Europe, where it’s too
expensive and politically difficult to build new
refineries. …The bulked-up Jamnagar will be
able to move markets: Singapore traders expect a
drop in fuel prices as soon as it’s going full tilt.
Fortune, April 28, 2008
a. Explain this news clip’s claims that the global
market for gasoline is not perfectly
b. What barriers to entry might limit competition in this market and give a firm such as Reliance power to influence the market price?
8. Congress Examines Giant Airline Merger
Congress Wednesday examined a proposed $3.1
billion merger that would create the world’s 9160335_CH13_p287-314.qxd 312 6/22/09 9:05 AM Page 312 CHAPTER 13 Oligopoly largest carrier as critics of the deal warned it
could drive up the price of air travel. …
[Committee Chairman James Oberstar] said the
Delta-Northwest merger would discourage competition at major hubs, reduce service to customers and result in higher fares. Delta Chief
Executive Officer Richard Anderson said that the
merger would not limit competition because the
carriers primarily serve different geographic
regions. … About a dozen witnesses scheduled to
testify before the House Subcommittee on
Transportation and Infrastructure were likely to
focus on whether a merger between Delta Air
Lines and Northwest Airlines would benefit consumers by lowering prices through cost savings,
or harm them by reducing competition.
CNN, May 14, 2008
a. Explain how this airline merger might increase
air travel prices.
b. Explain how this airline merger might lower
air travel production costs.
c. Explain how cost savings might get passed on
to travelers and might boost producers profits.
Which do you predict will happen from this
airline merger and why?
d. Explain the guidelines that the Federal Trade
Commission uses to evaluate mergers and why
it might permit or block this merger.
9. U.K. Price Cuts Seen for Apple’s New iPhone
AT&T is planning to sell the new iPhone this
summer for around $200. … The new iPhone
would be available later this month in the U.S.
through Apple and AT&T. AT&T pricing strategy calls for a $200 subsidy for customers who
sign two-year service contracts. …
The lower-priced phones in the U.S. … would
help juice the iPhone’s sales volume and give the
telcos an attractive weapon to win new subscribers.
In the United States, AT&T says it pulls in an
average of $95 a month from each iPhone customer, nearly twice the average monthly bill of
its conventional cell phone user. AT&T has a
revenue-sharing agreement with Apple that
requires it to give Apple as much as 25% of its
iPhone customers’ monthly payments.
Subsidies are a widespread pricing practice in the
United States and overseas. In exchange for a
cheaper phone, customers are locked in to a carrier for a year or two. It’s a small investment
by the telcos for a large return. … After giving
Apple its cut of the revenue, the remaining take
for AT&T is between $70 and $75 a month per
iPhone user, totaling more than $1,700 over the
life of the two-year contract. …
Fortune, June 2, 2008
a. How does this arrangement between AT&T
and Apple regarding the iPhone affect competition in the market for cell phone service?
b. Does this arrangement between AT&T and
Apple regarding the iPhone violate U.S. antitrust laws? Explain.
10. Starbucks Sued for Trying to Sink Competition
An independent coffee shop owner filed a lawsuit
against Starbucks Corp. Monday, charging the
coffee house giant with using anti-competitive
tactics to rid itself of competition. The suit …
was filed … by Penny Stafford, owner of the
Seattle-based Belvi Coffee and Tea Exchange Inc.
The lawsuit contends that Starbucks exploited its
monopoly power in the specialty retail coffee
market through such predatory practices as offering to pay leases that exceeded market value if
the building owner would refuse to allow competitors from occupying the same building.
Stafford says Starbucks also used methods such
as having employees offer free drink samples in
front of her store to lure away customers, which
she says ultimately forced her to close her
store. … The suit contends the world’s largest
specialty coffee retailer also used other predatory tactics nationwide including offering to
buy out competitors at below-market prices and
threatening to open nearby stores if the offer is
CNN, September 26, 2006
a. Explain how Starbucks is alleged to have violated U.S. antitrust laws in Seattle.
b. Explain why it is unlikely that Starbucks
might use predatory pricing to permanently
drive out competition.
c. What information would you need that is not
provided in the news clip to decide whether
Starbucks had practiced predatory pricing?
d. Draw a graph of the situation facing Belvi
Coffee and Tea Exchange Inc. when Penny
Stafford closed the firm. 9160335_CH13_p287-314.qxd 6/22/09 9:05 AM Page 313 Problems and Applications 11. Oil Trading Probe May Uncover Manipulation
Amid soaring oil prices … the government
Thursday announced a wide ranging probe into
oil price manipulation. … The CFTC
[Commodity Futures Trade Commission] is
looking into … manipulation of the physical oil
market … by commercial players who might literally withhold oil from the market in an attempt
to drive prices higher.
The CFTC has found evidence of this in the
past. … Haigh [a former economist at CFTC]
thinks it’s likely CFTC will find evidence of this
again given that the agency has been investigating
for six months and has now chosen to make it public. But he stressed that a single player acting alone
would in all likelihood not have a huge influence
on prices. “It’s difficult to imagine a price run-up
of $90 to $135 being done by one entity,” he said.
CNN, May 30, 2008
a. What type of market does former CFTC
economist Haigh imply best describes the
U.S. oil market?
b. Is economist Haigh’s comment consistent with
the predictions of the kinked demand curve
model of oligopoly? Explain.
c. Is economist Haigh’s comment consistent with
the predictions of the dominant firm oligopoly model? Explain.
d. Is economist Haigh’s comment consistent with
the predictions of any model of oligopoly?
12. Bud and Wise are the only two producers of
aniseed beer, a New Age product designed to displace root beer. Bud and Wise are trying to figure
out how much of this new beer to produce. They
know that if they both limit production to
10,000 gallons a day, they will make the maximum attainable joint profit of $200,000 a day—
$100,000 a day each. They also know that if
either of them produces 20,000 gallons a day
while the other produces 10,000 a day, the one
that produces 20,000 gallons will make an economic profit of $150,000 and the one that sticks
with 10,000 gallons will incur an economic loss
of $50,000. Each also knows that if they both
increase production to 20,000 gallons a day, they
will both make zero economic profit.
a. Construct a payoff matrix for the game that
Bud and Wise must play. 313 b. Find the Nash equilibrium of the game that
Bud and Wise play.
c. What is the equilibrium of the game if Bud
and Wise play it repeatedly?
13. Gadgets for Sale … or Not
How come some gadgets, like the iPod, cost the
same no matter where you shop? … No, the
answer isn’t that Apple illegally manages prices.
In reality, Steve Jobs and Co. use an accepted, if
controversial, tactic, a retail strategy called minimum advertised price, to discourage resellers
The minimum advertised price, or MAP, is the
absolute lowest price retailers are allowed to
advertise a product for. MAP is usually enforced
through marketing subsidies offered by a manufacturer to its resellers. If a retailer keeps prices at
or above the minimum advertised price, then a
manufacturer like Apple will give them money to
help advertise. If a store’s price dips too low, on
the other hand, the manufacturer can withdraw
these advertising subsidies. …
Stable prices are important to the company,
because it’s a manufacturer and a retailer (both
online and through its chain of Apple Stores). If
Apple resellers dropped prices on iPods and
iMacs—selling at or below cost to get customers
in the door, or as a way to cross-sell stuff like
software or iPod skins—they could squeeze the
Apple Stores out of their own markets.
There is a downside to all that stability, however.
By limiting how low sellers can go, MAP keeps
prices artificially high (or at least higher than
they might otherwise be with unfettered price
Slate, December 22, 2006
a. Describe the practice of resale price maintenance that violates the Sherman Act.
b. Describe the MAP strategy used by iPod and
explain how it differs from a resale price maintenance agreement that would violate the
c. Why might the MAP strategy be against the
d. Why might the MAP strategy benefit the consumer?
e. What is the bottom line on the MAP strategy:
does it benefit the consumer or only the
producer? 9160335_CH13_p287-314.qxd 314 6/22/09 9:05 AM Page 314 CHAPTER 13 Oligopoly 14. Asian Rice Exporters to Discuss Cartel
Rice-exporting nations planned to discuss a proposed cartel to control the price of the staple
food. … Rice exporters Thailand, Cambodia,
Laos and Myanmar planned to meet Tuesday to
discuss a proposal by Thailand, the world’s
largest rice exporter, that they form a cartel.
Ahead of the meeting … the countries sought to
assuage concerns that they might force up prices
by limiting supplies.
Unlike the Organization of Petroleum Exporting
Countries, the purpose of the rice cartel would
be “to contribute to ensuring food stability, not
just in an individual country but also to address
food shortages in the region and the world,”
Cambodian Prime Minister Hun Sen said
“We shall not hoard (rice) and raise prices when
there are shortages,” Hun Sen said. The
Philippines wasn’t convinced.
“It is a bad idea. … It will create an oligopoly
and it’s against humanity,” Edgardo Angara,
chairman of the Philippine Senate’s Committee
on Agriculture, said Friday, adding that the cartel
could price the grain out of reach for “millions
and millions of people.”
CNN, May 6, 2008
a. Assuming the rice-exporting nations become a
profit-maximizing colluding oligopoly, explain
how they would influence the global market
for rice and the world price of rice.
b. Assuming the rice-exporting nations become a
profit-maximizing colluding oligopoly, draw a
graph to illustrate their influence on the global
market for rice.
c. Even in the absence of international antitrust
laws, why might it be difficult for this cartel to
successfully collude? Use the ideas of game
theory to explain.
15. An Energy Drink with a Monster of a Stock
The $5.7 billion energy-drink category, in which
Monster holds the No. 2 position behind industry leader Red Bull, has slowed down as copycat
brands jostle for shelf space—and the attention
of teen consumers. … Over the past five years
Red Bull’s market share in dollar terms has gone
from 91 percent to well under 50 percent … and
much of that loss has been Monster’s gain.
Fortune, December 25, 2006 a. Describe the structure of the energy-drink
market. How has that structure changed over
the past few years?
b. Explain the various difficulties Monster and
Red Bull would have if they attempted to collude and charge monopoly prices for energy
16. Suppose that Firefox and Microsoft each develop
their own versions of an amazing new Web
browser that allows advertisers to target consumers with great precision. Also, the new
browser is easier and more fun to use than existing browsers. Each firm is trying to decide
whether to sell the browser or to give it away.
What are the likely benefits from each action?
Which action is likely to occur?
17. Why do Coca-Cola and PepsiCo spend huge
amounts on advertising? Do they benefit? Does
the consumer benefit? Explain your answer by
constructing a game to illustrate the choices
Coca-Cola and PepsiCo make.
18. Microsoft with Xbox 360, Nintendo with Wii,
and Sony with PlayStation 3 are slugging it out
in the market for the latest generation of video
games consoles. Xbox 360 was the first to market; Wii has the lowest price; PS3 uses the most
advanced technology and has the highest price.
a. Thinking of the competition among these
firms in the market for consoles as a game, describe the firms’ strategies concerning design,
marketing, and price.
b. What, based on the information provided,
turned out to be the equilibrium of the game?
c. Can you think of reasons why the three consoles are so different?
19. Study Reading Between the Lines on pp. 362–363
and then answer the following questions.
a. What were the strategies of Dell and HP in
2000 and in 2006?
b. Why, according to the news article, was Dell
having a harder time in 2006 than it had in
c. Why wouldn’t HP launch its new product and
marketing campaign and cut its price?
d. What do you think Dell must do to restore its
place as market leader?
e. How would you describe the global market for
PCs? Is it an example of oligopoly or monopolistic competition? ...
View Full Document
This note was uploaded on 02/07/2010 for the course ECON 251 taught by Professor Blanchard during the Fall '08 term at Purdue University-West Lafayette.
- Fall '08