0% found this document useful (0 votes)
4 views9 pages

Econ 487 Problem Set 7 Solutions

The document consists of a problem set for an economics course, focusing on utility functions, risk aversion, and game theory scenarios. It includes questions about expected utility calculations, risk-sharing agreements, and Nash equilibria in various games. Solutions to the problems are provided, demonstrating the application of economic theories and concepts.

Uploaded by

Ibrahima Diallo
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
4 views9 pages

Econ 487 Problem Set 7 Solutions

The document consists of a problem set for an economics course, focusing on utility functions, risk aversion, and game theory scenarios. It includes questions about expected utility calculations, risk-sharing agreements, and Nash equilibria in various games. Solutions to the problems are provided, demonstrating the application of economic theories and concepts.

Uploaded by

Ibrahima Diallo
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Econ 487

Problem Set 7 (Chapter 9)

1. Does the utility function given by u(money) = (money)0.3 show risk aversion,
lovingness or neutrality?
2.

3. Consider a variation of the scenario discussed in question 2 above in which now


not only you face the risky prospect but also your neighbor faces the risky
prospect, with your luck and your neighbor’s luck being completely opposite.
That is, with probability 0.6 you may get $160,000 but your neighbor in that case
gets $40,000, and with probability 0.4 you may get $40,000 but your neighbor in
that case gets $160,000. Assuming both of your utility functions are still given by
the square root function, does a risk-sharing scheme in which you and your
neighbor share your combined income equally make both you and your neighbor
better-off (compared to no such sharing) ?
4. S2
5. S3 (ignore part f)
6. S4
7. Nature determines whether the payoffs are as in Game 1 or as in Game 2 below,
each game being equally likely. Player 1 learns whether nature has drawn Game 1
or Game 2, but player 2 does not. Find the pure-strategy NE.
Player 2 Player 2

L R L R
Player 1 T 10, 10 20, 10 Player 1 T 10, 10 0, 0
B 0, 40 30, 10 B 20, 0 10, 30

GAME 1 1 GAME 2
8. Consider the following variation of Tudor-Fordor example considered in class.
Find the equilibrium in this game. Will it be a separating or pooling equilibrium?
149, − 27

€ 180, 0
8, 5


€ 30, 0
28, 5


€ 50, 0

9. S9 (in part d, you can calculate the expected payoff in the way we did in Chap 7).
10. S10 (ignore parts f and g; in part e, only find €the equilibria but don’t worry about
answering whether it’s a separating, pooling or semiseparating).

Solutions:

1. Make up a lottery: $100 with probability 0.5 and $400 with probability 0.5. (As
we said in class any lottery is fine.) Now, calculate the expected payoff (EP):

EP = 0.5(100) + 0.5(400) = 250.

EU of the lottery = 0.5(100)0.3 + 0.5(400)0.3 = 5.01

u(EP) = (250)0.3 = 5.24

So, a person with this utility function prefers the EP of a lottery over the lottery it
self à risk averse.

2. (a) In your bad time, to have $100k as the questions say, requires that your
neighbor gives you $60k in your bad time. So, this questions sets y = $60k. So, there is
just x in this problem.

$160k $100k $(160k – x) $(100k + x)


$40k $100k $100k $40k
?? ?? ?? ??

2
Before the payments are made, your neighbor’s EU = √100,000 = 316.23.

The lowest value of x such that your neighbor prefers to enter the contract will be the x
for which his expected utility for entering the contract is equal to his utility for not
entering the contract:

0.6 * √(100,000 + x) + 0.4 * √40,000 = 316.23

Þ √(100,000 + x) = (316.23 – 0.4 * √40,000) / 0.6

Þ x = [(√100,000 – 0.4 * √40,000) / 0.6]2 – 100,000 ≈ 55,009.8818

Rounding down to $55,009.88 would make your neighbor very slightly prefer not
entering the contract, so the minimum x that your neighbor will agree to is $55,009.89.

(b) Here we are looking for the level of x where you are indifferent between getting
insurance (where you pay x in a good year and receive 60,000 in a bad year) and not
getting insurance. That is, we’re looking for the x for which your expected utility with the
insurance is equal to your expected utility without the insurance:

0.6 * √(160,000 – x) + 0.4 * √100,000 = 0.6 * √160,000 + 0.4 * √40,000

Þ 0.6 * √(160,000 – x) + 0.4 * √100,000 = 320

Þ √(160,000 – x) = (320 – 0.4 * √100,000) / 0.6

Þ x = 160,000 – [(320 – 0.4 * √100,000) / 0.6]2 ≈ 55,984.1891

Rounding up and paying $55,984.19 would make you very slightly prefer not having
insurance. The highest x you would be willing to pay in a good year and still (barely)
prefer to have insurance is $55,984.18.

3. This is what we find:

$160k $40k $100k $100k


$40k $160k $100k $100k
?? ?? ?? ??

So, before sharing, your EU = 0.6 * √160,000 + 0.4 * √40,000 = 320

Neighbor’s EU = 0.6 * √40,000 + 0.4 * √160,000 = 280

After sharing, your EU = √100,000 = 316.23

3
Neighbor’s EU = √100,000 = 316.23

So, risk-sharing is making your neighbor better-off but not you.

4. (a) Since the payoff from not standing in line is 0 for both a student and a
homeless person, to achieve separation, t must be such that:

0 > 10 – t2 / 160 and 0 < 10 – t2 / 320

Þ t2 > 1,600 and t2 < 3,200

Þ t > 40 and t < 56.57

The minimum wait time that achieves separation is the smallest moment longer than forty
minutes.

(b) The expected benefit from standing in line for a college student is:

0.5*(10) + 0.5*(–5) = 2.5.

So, t must now satisfy:

0 > 2.5 – t2 / 160 and 0 < 10 – t2 / 320

Þ t2 > 400 and t2 < 3200

Þ t > 20 and t < 56.57

The minimum wait time that achieves separation is now a shade more than twenty
minutes. The partial identification of college students reduces the minimum wait time
required to achieve separation. When the charity has more information about the
patrons—even if only partial information—this allows it to distinguish between the two
types by means of a less stringent test.

5. (a) Buyers expect a random Citrus to be an orange with probability f = 0.6 and
a lemon with probability 0.4. Risk-neutral buyers are then willing to pay up to:

0.6 * $18,000 + 0.4 * $8,000 = $14,000

(b) The willingness to accept (WTA) of owners of oranges is $12,500. Since the
willingness to pay (WTP) of buyers is $14,000 for a random Citrus, sellers will be willing
to sell and buyers will be willing to buy for any price in the range [$12,500, $14,000].
Thus, there will indeed be a market for oranges.

(c) If f = 0.2, risk-neutral buyers will be willing to pay up to:

0.2 * $18,000 + 0.8 * $8,000 = $10,000

4
(d) There will not be a market for oranges when f = 0.2. The probability of a random
Citrus being an orange is so low that a buyer would only be willing to pay up to $10,000,
but owners of oranges require at least $12,500. No oranges will be sold.

(e) The minimum f such that oranges are sold is the f that satisfies:

fthreshold * $18,000 + (1 – fthreshold) * $8,000 = $12,500

Þ $10,000 * fthreshold = $4,500

Þ fthreshold = 0.45.

6. A competent electrician’s payoff after obtaining the signal (certification) is √100


– C; in the absence of the signal, this electrician’s payoff is √25. The competent
electrician will signal as long as √100 – C ³ √25, or as long as 10 – C ³ 5, or as long as 5
³ C. The incompetent electrician earns only √100 – 2C after certification versus √25
without certification. He signals only if √100 – 2C ≤ √25, or as long as 10 – 2C ≤ 5, or as
long as 5/2 ≤ C. The range of values of C for which the competent electrician signals but
the incompetent electrician does not is 5 ³ C ³ 2.5. If the program is restricted to an
integer number of months, 5 ³ C ³ 3.

7. Before we write down the normal form to find out the equilibrium, we need to
figure out how many strategies each player has. Since player 1 learns which game will be
chosen, he can make is strategy choice contingent on which game is played. So, he has
four strategies:

TT (choose T in game 1 and T in game 2).


TB (chose T in game 1 and B in game 2).
BT (choose B in game 1 and T in game 2).
BB (choose B in game 1 and B in game 2).

But player 2 does not know which game will be played. So, to him, it’s always just a
choice between L and R. So, player 2 has these two strategies, L and R.

5
Now, the strategic form of the game is:

Player 2

L R
0.5(10) + 0.5(10) = 10 , 0.5(20) + 0.5(0) = 10 ,
TT 0.5(10) + 0.5(10) = 10
0.5(10) + 0.5(0) = 5

0.5(10) + 0.5(20) = 15 , 0.5(20) + 0.5(10) = 15 ,


€ TB 0.5(10) + 0.5(0) = 5€ 0.5(10) + 0.5(30) = 20

Player 1 € 0.5(0) + 0.5(10) = 5 , 0.5(30) + 0.5(0) = 15 ,
€ BT 0.5(40) + 0.5(10) = €
25
0.5(10) + 0.5(0) = 5

€ 0.5(0) + 0.5(20) = 10 , 0.5(30) + 0.5(10) = 20 ,
€ BB €
0.5(40) + 0.5(0) = 20 0.5(10) + 0.5(30) = 20


€ €
Now, let’s find € € in the matrix above). So, (BB, R) is the
out the best responses (underlined
only pure-strategy NE.

8. The strategic form of the game is:

Fordor

II IO OI OO

LL 0.4(149) + 0.6(8) = 64.4, 0.4(149) + 0.6(8) = 64.4, 0.4(180) + 0.6(30) = 90, 0.4(180) + 0.6(30) = 90,
0.4(– 27) + 0.6(5) = –7.8 0.4(– 27) + 0.6(5) = –7.8 0.4(0) + 0.6(0) = 0 0.4(0) + 0.6(0) = 0

Tudor 0.4(149) + 0.6(28) = 76.4, 0.4(149) + 0.6(50) = 89.6, 0.4(180) + 0.6(28) = 88.8, 0.4(180) + 0.6(50) = 102,
LH
0.4(– 27) + 0.6(5) = – 7.8 0.4(– 27) + 0.6(0) = – 10.8 0.4(0) + 0.6(5) = 3 0.4(0) + 0.6(0) = 0

After finding the best responses (as shown by the underlines), the unique pure-
strategy Nash equilibrium of this game is (LL, OI). This is a pooling equilibrium because
both a low-cost Tudor and a high-cost Tudor send the same signal—setting a low price.

9. (a) The extensive form of Stripped-down Poker is:

6
(b) Felix has two actions at each of two nodes, so he has 4 strategies:

BB (Bet if K, Bet if Q)
BF (Bet if K, Fold if Q)
FB (Fold if K, Bet if Q)
FF (Fold if K, Fold if Q)

Oscar has two actions at a single information set, so he has two strategies.

(c) The strategic form is:

Oscar
Call Fold
0.5(2)+ 0.5(-2) = 0, 0.5(1)+ 0.5(1) = 1,
BB
0.5(-2)+ 0.5(2) = 0 0.5(-1)+ 0.5(-1) = –1
Felix 0.5(2)+ 0.5(-1) = 0.5, 0.5(1)+ 0.5(-1) = 0,
BF
0.5(-2)+ 0.5(1) = –0.5 0.5(-1)+ 0.5(1) = 0
0.5(-1)+ 0.5(-2) = –1.5, 0.5(-1)+ 0.5(1) = 0,
FB
0.5(1)+ 0.5(2) = 1.5 0.5(1)+ 0.5(-1) = 0
0.5(-1)+ 0.5(-1) = –1, 0.5(-1)+ 0.5(-1) = –1,
FF
0.5(1)+ 0.5(1) = 1 0.5(1)+ 0.5(1) = 1

7
(d) FB is dominated by BB, and FF is dominated by BF. The reduced game table,
with best responses underlined, is:

Oscar
Call Fold
Felix
BB 0, 0 1, – 1
BF 0.5, – 0.5 0, 0

After finding the best responses (as shown by the underlines), we see that the game has
no equilibrium in pure strategies. But it has a mixed-strategy Nash equilibrium where
Felix plays BB with probability 1/3 and Oscar Calls with probability 2/3. (Solve using the
technique we learned in Chap 7 to check these values.)

The expected payoff to Felix in equilibrium is: (2/3)(0) + (1/3)(1) = 1/3.

10. (a) If Wanda has a good year, reports a low income, and is then audited, she
will have to pay her tax liability of $5,000 and lose $1,000 worth of her time, so her
payoff is – $6000 in this outcome. The IRS receives $5,000 in tax from Wanda, but has to
pay $1,000 in administrative costs, so its payoff in this outcome is $4,000.

(b) Wanda has an incentive to bluff, which in this case would consist of her
reporting a low income in a good year. If she reports a low income, the IRS doesn’t know
if it is low because she had a bad year or because she had a good year and is bluffing. If
they decide not to audit when Wanda bluffs, she stands to save a considerable amount of
money.

(c) The extensive form is:

8
(d) Wanda has two actions at each of two nodes, so she has 4 strategies:

LL (L if G, L if B)
LH (L if G, H if B)
HL (H if G, L if B)
HH (H if G, H if B)

The IRS has two actions at a single information set, so it has two strategies.

(e) The strategic form is:

IRS
A N
LL 0.6(– 6,000) – 0.4(1,000) = – 4,000,
0, 0
0.6(4,000) – 0.4(1,000) = 2,000
Wanda LH 0.6(– 6,000) – 0.4(5,000) = –5,600, 0.6(0) – 0.4(5,000 = – 2,000,
0.6(4,000) + 0.4(5,000) = 4,400 0.6(0) + 0.4(5,000) = 2,000
HL 0.6(– 5,000) – 0.4(1,000) = –3,400, 0.6(–5,000) + 0.4(0) = –3,000,
0.6(5,000) – 0.4(1,000) = 2,600 0.6(5,000) + 0.4(0) = 3,000
HH – 5,000, 5,000 – 5,000, 5,000

Strategy LH is dominated by strategy LL, and strategy HH is dominated by both


LL and HL. The smaller version of the game table, with best responses underlined, is
thus:

IRS
A N
Wanda – 4,000, 2,000 0, 0
LL
HL – 3,400, 2,600 – 3,000, 3,000

After finding the best responses (as shown by the underlines), we see that there is
no pure-strategy Nash equilibrium. The mixed-strategy Nash equilibrium occurs when
Wanda plays LL with probability 1/6 and the IRS audits with probability 5/6. (Solve
using the technique we learned in Chap 7 to check these values.)

You might also like