SOLUTION TO SAMPLE EXAM 2

1. (10 points) Define Adverse Selection and Moral Hazard. How are they different? Give
examples of economic situations which suffer from adverse selection and moral hazard.
Adverse Selection: unobservable characteristic; Moral Hazard: unobserved ac-
tion.
This is not full answer: 1) you should say what adverse selection and moral
hazard actually are. I don’t do this here, b/c definitions are clear from studying
materials. Here I give only the essence of the definitions; 2) you need to provide
small examples of both.

2. (10 points) What is risk aversion? Give definition in words and provide a condition on
utility function that defines it.
Give some definition in words (see the textbook).
Between certain outcome and the lottery with the same expected value risk
averse person chooses sure outcome:

u( pA + (1 − p) B) > pu( A) + (1 − p)u( B)

3. (10 points) Give an example of a commitment device. What is credible and incredible
threat?
Doomsday device in the Dr. Strangelove movie is a perfect commitment device.
It creates credible threat because its creators cannot stop it if they are attacked.
Since the creators of Doomsday device removed the possibility of them stopping
it, the threat is credible and other party is not willing to attack. Incredible threat
by a player is a situation such that when it actually gets to threatening the other
party it is not in player’s interests to do that. So, threatened party is not “afraid”
of it.

4. (10 points) Find Nash Equilibria of the following game:

A B C
N 3,2 2,1 1,5
W 1,4 5,2 1,3
S 4,5 3,3 2,4
E 1,4 1,5 5,0

1
 Answer: (S, A). See sample exam 1 for more comments about answering such
questions.

5. The economy consists of two firms called 1 and 2 and two goods called G1 and G2 . Firm
i (i = 1, 2) produces only good Gi . The goods are substitutes, so the inverse demands
for goods G1 and G2 are given by

P1 (q1 , q2 ) = a − q1 − kq2
P2 (q1 , q2 ) = a − q2 − kq1

where q1 and q2 are the amounts of goods G1 and G2 produced by firms 1 and 2 and
1 ≤ k < 1.1 is a constant. The marginal costs for both firms are c. Firm i’s (i = 1, 2)
profit is therefore Pi (q1 , q2 )qi − cqi .

a) (10 points) Suppose that firms simultaneously choose quantities. Find a Nash
Equilibrium.
When firms simultaneously choose quantities they maximize their profit
taking the other firm’s quantity as given. Firm 1 solves:

max q1 ( a − q1 − kq2 ) − cq1

Taking derivative with respect to q1 we get:

a − 2q1 − kq2 − c = 0

which results after rearrangement in the best response of firm 1 to the

a − kq2 − c
q1 =
2
Firm 2 does similar thing. It solves:

max q2 ( a − q2 − kq1 ) − cq2

After differentiation and rearrangement we get the best response of firm 2

a − kq1 − c
q2 =
2

2
 Nash Equilibrium are the quantities of both firms such that these quantities
are best responses to each other. So the two best response equations above
should hold simultaneously. Plug BR of firm 2 to the BR of firm 1 and get

a−kq1 −c
a−k 2 −c
q1 =
2
Solve this for q1 and find

a−c
q1 =
2+k
Then plug this into BR of firm 2 and find

a−c
q2 =
2+k
a−c
So q1 = q2 = 2+ k is NE.
b) (10 points) Now suppose that first firm 1 chooses its quantity, then firm 2 ob-
serves the choice of firm 1 and also chooses quantity. Use backward induction to
find equilibrium outcome (if you get a messy formula with parameters a, k and c
do not spend time simplifying, leave it as is).
Here firms move one after another and firm 1 can predict the response of
firm 2 to any quantity that firm 1 chooses. BR of firm 2 to some q1 is the
same as in previous part, namely:

a − kq1 − c
q2 =
2
Now firm 1 takes this into account and solves:

a − kq1 − c
max q1 ( a − q1 − k ) − cq1
q1 2
Taking derivative with respect to q1 gives:

a−c
a − 2q1 − k + k 2 q1 − c = 0
2
Rearranging we get

( a − c)(2 − k)
q1 =
2(2 − k 2 )

3
 This is the optimal thing to do for firm 1 anticipating the response of firm
2. Firm 2 will thus choose

( a−c)(2−k)
a−k 2(2− k 2 )
−c
q2 =
2
Simplification is not necessary.

6. Consider Centipede Game (players are 1 and 2, the top payoff is of player 1, the bottom
- of player 2):

1 B 2 D 1 F 2 H 1 J 2 L
6
A C E G I K 6

5 2 6 3 7 4
0 5 2 6 3 8

a) (10 points) Find equilibrium outcome using backward induction. State equilib-
rium strategies for both players.
We use Backward Induction to find equilibrium outcome. We start from
the last move of player 2, the choice between K and L. Player 2 chooses
K (8 > 6). Now before this move Player 1 chooses between I and J, he
anticipates player 2 to go K, so he chooses I (7 > 4). Proceed in the same
way and find that equilibrium is: Player 1 goes A, E, I and player 2 goes C,
G, K.
b) (10 points) Suppose centipede game is repeated infinitely many times. For which
discounting factor δ can the outcome (6, 6) be sustained as equilibrium payoff in
each period? Find Nash equilibrium of the repeated game that does it and prove
that it is indeed an equilibrium (Hint: make sure your δ works for both players).
Consider the following strategy for Player 1: go B, F, J as long as Player
2 goes D, H, L; If any other action is observed go A, E, I forever after re-
gardless of what happens. Player’s 2 strategy is: go D, H, L as long as
Player 1 goes B, F, J; If any other action is observed go C, G, K forever after
regardless of what happens.
On the equilibrium path both players get discounted payoff of 6/(1 − δ).
We should check for which δ no one wants to deviate. Consider the best

4
 deviation of player 1: his maximal payoff in the game is 7, so the best
deviation given the above strategy of player 2 will give him:

δ
7 + δ5 + δ2 5 + δ3 5 + ... = 7 + 5
1−δ
So player 1 does not want to deviate if

δ
6/(1 − δ) ≥ 7 + 5
1−δ
which gives

1
δ≥
2
For player 2 the best payoff in the game is 8. So the best possible deviation
for player 2 given the above strategy of player 1 will give him

8 + δ0 + δ2 0 + δ3 0 + ... = 8

So player 2 does not want to deviate when

6(1 − δ ) ≥ 8

which gives

1
δ≥
4
Since 14 < 12 we conclude that if both players have δ ≥ 1
2 the above strate-
gies are equilibrium.

7. Consider health insurance company and a person (say, Bob) who wants to be insured.
Bob gets salary wh if he is healthy. If he gets sick he has to pay medical expenses, so
after that he is left with ws (wh > ws ). Bob is risk averse, his utility of money is
√
U ( x ) = x. Also, Bob gets sick with probability π and insurance company knows
that. Suppose insurance company proposes Bob to pay some amount p for insurance
and in return it guarantees to fully insure Bob in case of sickness.

a) (5 points) What does it mean for Bob to be fully insured? Give a definition.
This means that insurance company covers all medical expenses: Bob has
certain outcome of wh regardless of whether he is sick or not.

5
b) (5 points) What is the maximal price p that Bob is willing to pay to the insurance
company for full insurance? Provide a formula, do not simplify, leave it as is.
Maximal p solves
√ √ √
w h − p ≥ (1 − π ) w h + π w s

c) (5 points) What is the smallest price p that insurance company is willing to ac-
cept? Provide a formula, do not simplify, leave it as is.
The smallest price p insurance company is willing to accept is equal to its
cost:
p = π ( wh − ws )

d) (5 points) Suppose insurance company is not sure what are the chances of Bob
getting sick. Insurance company thinks that π = π1 with probability γ and
π = π2 , where π1 > π2 , with probability 1 − γ. What is the smallest price
insurance company is willing to accept in this case?
Now the smallest price is the expected cost:

p = γπ1 (wh − ws ) + (1 − γ)π2 (wh − ws )