Chapter 10 - Game Theory: Inside Oligopoly

Chapter 10: Answers to Questions and Problems

1.
a. Neither player has a dominant strategy.
b. Given the worst possible scenario, the highest guaranteed payoff for Player 1 is B
and the highest guaranteed payoff for Player 2 is E.
c. Nash equilibrium states, given the strategies of other players, no player can
improve their payoff by unilaterally changing their own strategy.

2.
a.

Player 2

Strategy A B

Player 1 A $400, $400 $100, $600

B $600, $100 $200, $200

b. B is dominant for each player.

c. (B, B).
d. Joint payoffs from (A, A) > joint payoffs from (A, B) = joint payoffs from (B, A)
> joint payoffs from (B, B).
e. No; each firm’s dominant strategy is B. Therefore, since this is a one-shot game,
each player would have an incentive to cheat on any collusive arrangement.

3.
a. Player 1’s optimal strategy is A. Player 1 does not have a dominant strategy.
However, by putting herself in her rival’s shoes, Player 1 should anticipate that
Player 2 will choose E (since E is Player 2’s dominant strategy). Player 1’s best
response to E is A.
b. Player 1’s equilibrium payoff is 18.

10-1
Chapter 10 - Game Theory: Inside Oligopoly

4.
a. (A, C).
b. No.
c. If firms adopt the trigger strategies outlined in the text, higher payoffs can be
 Cheat   Coop 1
achieved if  . Here, πCheat = 60, πCoop = 50, πN = 10, and the
 Coop
 N
i
 Cheat   Coop 60  50 1 1 1
interest rate is i = .05. Since    0.25 <   20
 Coop
 N
50  10 4 i .05
each firm can indeed earn a payoff of 50 via the trigger strategies.
d. Yes. With θ sufficiently low, this resembles the infinitely repeated game.

5.
a. x > 4.
b. x < 5.
c. x < 5.

6.
a. See the accompanying figure.

($0, $25)
Right

Right ($20, $20)

Left 2

Left (-$5, $10)

b. ($0, $25) and ($20, $20).

c. ($20, $20) is the only subgame perfect equilibrium; the only reason ($0, $25) is a
Nash equilibrium is because Player 2 threatens to play left if 1 plays left. This
threat isn’t credible.

10-2
Chapter 10 - Game Theory: Inside Oligopoly

7.
a. Player 1 has two feasible strategies: A or B. Player 2 has four feasible strategies:
(1) W if A and Y if B; (2) X if A and Y if B; (3) W if A and Z if B; (4) X if A and
Z if B.
b. There are three Nash equilibria, listed as follows: 1) Player 1 plays A, and Player
2 plays W if A and Y if B, 2) Player 1 plays B, and Player 2 plays W if A and Z if
B, 3) Player 1 plays B, and Player 2 plays X if A and Z if B. The first equilibrium
results in payoffs of (60, 120). The last two equilibria each result in payoffs of
(100, 150).
c. The subgame perfect equilibrium is: Player 1 plays B, and Player 2 plays W if A
and Z if B. Here, Player 2 makes the optimal choice for him at each node of the
game, and makes only credible threats. These strategies result in payoffs of (100,
150).

8.
a. There are two Nash equilibrium outcomes: (25, 25) and (15, 15). The (15, 15)
equilibrium would seem most likely since the other equilibrium entails
considerable risk if the players don’t coordinate on the same equilibrium. More
specifically, the (15, 15) equilibrium involves each player playing his secure
strategy.
b. “A”. This would signal to player 2 that player 1 is going to use strategy A, and
therefore permit the players to coordinate on the equilibrium resulting in (25, 25).
c. Player 2 would choose X and player 1 would follow by choosing A. This is the
subgame perfect equilibrium, and would result in the outcome of (25,25).

9.
a. Both players have dominant strategies. Player 1’s dominant strategy is A and
Player 2’s dominant strategy is C.
b. Player 1’s secure strategy is A. Player 2’s secure strategy is C.
c. The Nash equilibrium is for Player 1 to play A and Player 2 to play C, resulting in
payoffs of (-10, -10).

10. Player 1’s trigger strategy to sustain profit of $140 is to play B until Player 2 deviates
from strategy D, then play A forever after. Player 2’s trigger strategy is to play D
until Player 1 deviates from strategy B, then play C forever after.

10-3
Chapter 10 - Game Theory: Inside Oligopoly

11. The normal form game looks like this:

Kmart
Strategy Sale Price Regular Price
Target Sale Price $2, 2 $7, $4
Regular Price $4, $7 $4, $4

Notice that there are two Nash equilibria: (Sale, Regular) and (Regular, Sale) with
profits of ($7, 4) and ($4, $7), respectively. Thus, there is not a clear-cut pricing
strategy for either firm. One mechanism that might solve this problem is to advertise
your sales on alternate weeks. Another mechanism might be to guarantee “everyday
low prices” (so that you effectively commit to always charge the sale price). In this
case, your rival’s best response would be to charge the regular price and your firm
would earn profits of $7 billion.

12. The normal form game looks like this:

Honda
Strategy Airbags No Airbags
Toyota Airbags $2.5, 2.5 $3,-$1.5
No Airbags -$1.5, $3 $1, $1

The dominant strategy, in this case, would be to offer airbags.

13. The extensive form game looks like this:

($200, $300)
Not Introduce

P
1
Price War ($100, $100)

Introduce C
2

Acquiesce ($227, $275)

Notice that Coca-Cola’s best response if Pepsi introduces is to acquiesce to earn $275
million rather than to start a price war and earn $100. Thus, while Coca-Cola might
threaten to start a price war in an attempt to keep you out of the market, this threat
isn’t credible; your best option is to introduce.

10-4
Chapter 10 - Game Theory: Inside Oligopoly

14. The savings from letting the union use its own pen and ink to craft the document are
most likely small compared to the advantage you would gain by making a take-it-or-
leave-it offer.

15. Since you know for certain that the game will end in 1 month, your optimal strategy
in the finitely repeated pricing game with a known endpoint is to reduce price (defect)
from the implicit collusive agreement between you and your rival.

16. The normal form of this game looks as follows:

Rival
Strategy: Price Low High
You Low $0, $0 $11,-$2
High -$2, $11 $6, $6

The one-shot Nash equilibrium is for both firms to charge a low price to earn zero
profits. Now suppose you and your rival compete year after year but there is a 60
percent chance the Highlander is discontinued. The profits of a firm that conforms to
the collusive strategy (high price) under the usual trigger strategies (firms agree to
charge the high price so long as no player deviated in the past, otherwise charge a low
$
price) are = $6 + $6(1 − 0.6) + $6(1 − 0.6) + ⋯ = . = $10 million. A
firm that cheats earns $11 million today and zero forever after. Since πCoop < πCheat,
the collusive outcome cannot be sustained as a Nash equilibrium, the collusive
outcome can be sustained as a Nash equilibrium.

17. The normal form game looks like this:

Rival

Advertise No Yes
Kellogg’s
No $12, $12 -$4, $52
Yes $52,-$4 $0, $0

Collusion is profitable under the usual trigger strategies if ≤ , or

$ $
$
= 3.33 ≤ . Thus, one requirement is for the interest rate to be less than 30
percent. Another requirement includes the ability of firms to monitor (observe)
potential deviations by rivals.

10-5
Chapter 10 - Game Theory: Inside Oligopoly

18. The normal form looks like this, where payoffs are the profits for each firm:

Baker
Price $10 $20

$5 9, 16 9,18
Argyle
$10 8, 16 8,18

Your optimal price is $5, since that strategy is a dominant strategy. You should not
invest the $2 million. This is because your profits do not depend on your rival’s
price. Therefore, there can be no payoff advantage to moving first.

19. Direct labor and direct materials are the only relevant costs, since they are variable
costs. Depreciation is a fixed (or sunk) cost, and is therefore irrelevant to the decision
(the firms’ fixed costs are $30,000 (since $30,000/250 = $120 and $30,000/500 =
$60. These later numbers are the reported unit depreciation costs). The firm’s
marginal cost (which equals its average variable cost in this case) is thus the sum of
unit labor and materials costs, or $35 + $25 = $60. The payoff matrix (normal form)
below shows the relevant payoffs (contributions) towards paying the $30,000 in fixed
costs for alternative levels of output by the two firms. The key is to note that if each
firm produces 250 units, total market output is 500 units and the price is $140. Each
firm’s contributions in this case are ($140 - $60) x 250 = $20,000. If one firm
produces 250 units and the other firm produces 500 units, the market price is $110.
In this case, the firm’s contributions are ($110 - $60) x 250 = $12,500 and ($110 -
$60) x 500 = $25,000. If each firm produces 500 units, the market price is $95, and
the contributions of each firm are ($95 - $60) x 500 = $17,500. The payoff matrix
(normal form) below shows the relevant contributions toward paying the fixed costs
of $30,000.

NetWorks
Strategy 250 Units 500 Units
GearNet
250 Units $20000, $20000 $12500, $25000
500 Units $25000, $12500 $17500, $17500

Each firm’s dominant strategy in a one-shot game is to produce 500 units. In

equilibrium each firm contributes $17,500 toward its $30,000 in fixed costs.

10-6
Chapter 10 - Game Theory: Inside Oligopoly

20. The normal-form representation of this game is depicted in the following payoff
matrix.

T-Mobile
Strategies CDMA GSM
Qualcomm CDMA $16 b, $12.9 b $12.3 b, $8.6 b
GSM $14.7 b, $7.4 b $13.5 b, $18.7 b

There are two Nash equilibria to this coordination game: (1) Qualcomm and T-
Mobile adopt the CDMA technology and (2) Qualcomm and T-Mobile adopt the
GSM technology. There are many ways to solve multiplicity of equilibria in this
coordination problem. As the book points out, the firms could “talk” to each and
agree on one technology. Alternatively, the developing country’s government could
announce which technology is to be used in the country.

21. The normal form of this game is contained in the following payoff matrix (in billions
of U.S. dollars).

Japan
Strategies Tariff No Tariff
U.S.
Tariff $49.1, $9.5 $52.5, $8.9
No tariff $48.2, $11.4 $50, $10

The Nash equilibrium is for the U.S. and Japan to each impose tariffs. However, both
countries achieve greater welfare by “agreeing” to impose no tariffs. The
sustainability of such an agreement to impose no tariffs is dependent upon the game
being repeated infinitely (or having sufficiently low probability of ending in a given
period), the countries using trigger strategies and the interest rate being sufficiently
low.

22. You should not recommend that the office manager invest more time monitoring. The
problem is not that she is monitoring too little. Rather, her monitoring activities and
strategies are predictable. Workers realize that once she leaves after the 9 a.m. check,
she is unlikely to return until 11 a.m. Recognizing this, workers know they will not
get caught “goofing off” (shirking). The manager’s best strategy is to randomize both
the timing and number of checks she does each day. That way, her monitoring is not
predictable and workers will respond by spending less time shirking.

10-7
Chapter 10 - Game Theory: Inside Oligopoly

23. If Congress passes the tariff, each firms gains $6 million in “extra” profit (= $30 / 5).
If your firm commits to not spending any money on lobbying, one or more of the
other firms in the industry would have an incentive to collectively spend $5 million
on lobbying. Under this scenario, your “optimal” profits are $6, compared to profits
of $1 million when you pay $5 million on lobbying. However, if your rivals knew
that you were willing to pay the entire lobbying bill, your threat is not credible and
your competitors would not be inclined to spend any money on lobbying. More
formally, this is a coordination game with multiple Nash equilibria. In one of the
equilibria, your firm spends nothing on lobbying and one or more competitors
collectively spend $5 million on lobbying such that the proposed tariff passes.
Another equilibrium occurs when your rival firms spend nothing on lobbying
activities and you pay the entire $5 million in lobbying expenses. The natural “focal
point” is for each firm to agree to spend $1 million on lobbying. This results in each
of the five firms earning a profit (net of lobbying costs) of $5 million.

10-8