Final - Answers
Final - Answers
Two firms with constant marginal costs serve two markets for two different goods. The
demand function for good 1 is Q1 = 20 − p1 + 0.5p2 , where p1 and p2 are the prices of good
1 and good 2, respectively. The demand function for good 2 is Q2 = 25 + p1 − p2 .
To answer this problem, you need not perform any calculations. (You certainly may if
you want, but the answers have to contain economic intuition).
The goods are substitutes, since cutting the price of the one reduces the
demand for the other and vice versa.
ii. (10 points) Suppose that we start from the competitive outcome. Will the firms
want to collude? What will happen to the prices if the firms collude?
Hint: Think about the externalities that firm 1 creates for firm 2 if firm 1, say,
increases the price of its product.
The firms will surely want to collude to eliminate the externality prob-
lem.
1
iii. (10 points) How would your answer to (ii) change if the demand functions were
Q1 = 20 − p1 − p2 and Q2 = 25 − p1 − p2 ?
Here the externality runs the other way: the price cuts firm 1 makes
are good news for firm 2 (the goods are now complements). Under
competition firms do not care about externalities and the prices will be
too high compared to the collusive outcome. Hence, the collusion now
brings about price decline. It happens because now the goods the firms
produce are complimentary, and their rivalry is detrimental.
Clearly, the firms will want to collude in this case as well, because the
externality is still around, even though it goes the other way.
No, they will not benefit from that, because the markets are completely
independent and are monopolized at the very start. You can call this
monopoly in two markets two different names or call it one name – it
will not change anything.
Sugar Kowalczyk enjoys money, M, and saxophone music, S. Her utility function is U =
S 2 + M and her budget constraint is pS + M = I.
i. (15 points) Derive Sugar’s demand for saxophone music and money. Is the law of
demand satisfied for Sugar’s demand for saxophone music? What is the economic
2
intuition behind your answer? What change, if any, should be made in the utility
function to make the consumer behave in line with the law of demand?
M US pS p
= ⇒S= ,
M UM pM 2
and it is the demand function for saxophone music.
To get the demand for money, substitute the demand for saxophone
music into the budget constraint: M = I − p2 /2
The demand function for S violates the law of demand, as the demand
for S increases with price.
The reason is that the marginal utility M US for this utility function
also increases with S (for standard utility functions it would decrease).
Clearly, Sugar is addicted to saxophone music and wants more and more
of it after she hears some, eventually ending up paying a high price, as
she actually confesses in the movie (‘Some Like It Hot’). Alas, there is
nothing irrational in it: if the price Sugar has to pay for the music is
high, she has to consume a lot and get addicted, otherwise she will have
to give up saxophone music (and saxophone players) altogether, and we
know from the movie she just cannot do it.
ii. (10 points) Is Sugar risk-averse with respect to random fluctuations in income? What
about fluctuations in the price of music?
Hint: Find the indirect utility function, i.e. express Sugar’s utility in terms of her
income and the price of music. Then determine if Sugar is risk-averse with respect
to fluctuations in I treating p as a constant, and vice versa, that is, would she rather
take a fixed income (price) or a gamble with the same mathematical expectation.
p p2 p2
x= , y=I− ⇒V =I−
2 2 4
3
The indirect utility function V is linear in I, so Sugar is risk-neutral
with respect to fluctuations in income. V is concave in p, so Sugar is
risk-averse with respect to fluctuations in the price of music.
Consider a monopoly which faces the demand curve P = 150 − 2Q and has MC=30.
Assume FC=0.
i. (5 points) What is the optimal quantity for the monopoly? What profit does it make
at this quantity?
ii. (5 points) What is the point elasticity of demand at the optimal quantity? What
P − MC
is the markup (defined as ) at the optimal quantity? What is the relation
P
between the two?
Notice that the demand function you are given (commonly referred to
as inverse demand function) expresses P in terms of Q, and to compute
elasticity you need it to do otherwise. So, the demand function equation
you need is
P dQ P P/2
Q = 75 − ⇒η=− · =
2 dP Q 75 − P/2
4
P − MC 90 − 30 2
Q = 30 ⇒ P = 150 − 2 · 30 = 90 ⇒ = =
P 90 3
90/2 3
Q = 30 ⇒ η = =
75 − 90/2 2
The optimal markup is the inverse of the demand elasticity at the opti-
mum point. The derivation in (iii) shows that it is always true.
Remark : Notice that you could take the elasticity of the inverse demand
dP Q
function straight away, defining it as ε = − · . Just remember that
dQ P
what you will get is not the demand elasticity – ε = 1/η.
iii. (10 points) Assuming your answer to (ii) is true in general, will a monopoly ever
produce on the inelastic portion of its demand curve? To get full credit, provide a
rigorous proof that it will or will not.
On the inelastic part of the demand curve the inverse of the elasticity
P − MC
is greater than 1, which would imply > 1 ⇒ M C < 0, which does
P
not make sense. So, the monopoly will never produce on the inelastic
part of the demand curve.
5
revenue is decreasing in quantity). It is also true that the elasticity of
demand is always greater than 1 to the left of this point.
M C = p ⇒ 150 − 2Q = 30 ⇒ Q = 60 ⇒ P = 30
v. (10 points) What is the gain/loss to the society from switching to the competitive
outcome?
Under competition, the firm gets zero profit, but CS is now equal to
the area of the triangle 150-B-C (and equal to SW). So, under competi-
tion SW=CS=0.5*(150-30)*60=3600, and the gain to the society from
switching to competition is 3600-2700=900. You may notice that the
area of triangle ABD is equal exactly 900 – it is the dead-weight loss
triangle.
Consider a firm that faces the demand curve P = 150 − Q and has M C = $20.
i. (5 points) If the firm is a monopoly, what is its optimal price and quantity? What
is the profit?
ii. (10 points) Suppose another firm enters the industry and the two firms compete a-la
Cournot. The entrant has M C = $40 and has to pay $500 to enter the industry.
6
Compute the optimal quantity for the entrant and the incumbent, their profits and
the price.
M RI = 150 − 2QI − QE = M CI = 20 2QI + QE = 130 QI = 50
⇒ ⇒ ⇒
M RE = 150 − QI − 2QE = M CE = 40 QI + 2QE = 110 QE = 30
⇒ p = $70 ⇒ ΠI = 50 · ($70 − $20) = $2500, ΠI = 30 · ($70 − $40) − $500 = $400
iii. (10 points) Suppose the incumbent can credibly commit to producing and selling 80
units of the good. Will the entrant still want to enter? Will the incumbent want to
commit to selling 80 units of the good?
M RE = 70 − 2QE ; M RE = M CE ⇒ 70 − 2QE = 40 ⇒ QE = 15 ⇒
Since the entrant’s profit is negative, it will not enter. Therefore, the
incumbent will want to commit to producing and selling 80 units.
iv. (15 points) Suppose the incumbent still commits to produce 80 units, but the entrant
does not have to pay anything to enter. Will it enter? What will its quantity choice
be? Are the quantity choices of the entrant and the incumbent a Nash equilibrium
in the Cournot game?
If there are no entry costs, the entrant will enter the industry – in (iii)
it makes a positive profit before entrance costs. It will still choose to
produce 15 units, because fixed costs do not influence pricing.
The quantity choices are not a Nash equilibrium. The entrant will
not want to deviate given the incumbent’s choice, because it already
has maximum profit given the choice. The incumbent, though, will be
tempted to revise the choice given that the entrant sticks to producing
15 units. The incumbent then faces the residual demand of P = 135−QI .
So,
7
⇒ p = $77.5 ⇒ ΠI = 57.5 · ($77.5 − $20) = $3306.25 > $2800
v. (25 points) Suppose the entrant pays nothing to enter and offers to form a cartel.
The cartel is supported by the usual trigger strategy with the threat to revert to
the Cournot equilibrium forever. Will the cartel be sustainable if they divide the
monopoly profit 50-50? What if they divide it 70-30 in favor of the incumbent? What
is the maximum and the minimum amount of money the incumbent can receive as
a cartel member so that the cartel is sustainable? Assume that the firms use the
discount rate of 15% to compute the present value of the future profits.
If the entrant deviates, it will just work on the residual demand after
the incumbent sells the monopoly output, P = 150 − 65 − Q = 85 − Q
8
not be able to change the output, but will be able to hold back the check
for $1237.5 it would have owed the entrant this period if the entrant
had been true to the cartel agreement. Then the entrant will never
deviate, because it loses both when it deviates and afterwards. In fact,
the incumbent can pay as few as 0.15 · (506.25 + 900/(0.15 · 1.15 ≈ 858.5 to
the entrant and still keep him in the cartel (i.e., out of the market).
Notice, by the way, that if the incumbent pays the entrant less than
$900, the entrant will actually want to be ”punished” by the Cournot
equilibrium, but will refrain from breaking the cartel agreement because
deviating means punishing itself quite severely.
Alternatively, we can assume that the incumbent is a bit slow and will
deliver to the entrant the 30% of its monopoly profit even in the period
the entrant deviates. Of course, the incumbent will not be able to send
the entrant $1267.5, rather, it will send 0.3 · 65 · (62.5 − 20) = 828.75, so
the deviating entrant will get $506.25 + $828.75 = $1335. The entrant now
compares 1267.5/0.15 = 8450 from being in the cartel and 1335 + 900/(0.15 ·
1.15) ≈ 6552.4 from deviating and chooses to stay in the cartel. The
minimum fraction of the monopoly profit the incumbent has to offer
the entrant to make it stay out solves
4225 900
x· = 506.25 + x · 65 · 42.5 + ⇒ x = 0.2253
0.15 0.15 · 1.15
So, the incumbent has to offer the entrant at least 0.2253 · 4225 = 951.9 to
keep it out of the market.
NB: In your exam, you could make any of the two assumptions and pro-
ceed relying on it. You did not need to discuss what happens under both
assumptions.
Consider a firm with MC=AC=1 trading with two buyers, whose demand functions are
Q1 = 5 − 2p1 and Q2 = 7 − 3p2 . The firm can distinguish the buyers.
9
i. (5 points) What is the optimal price-discrimination strategy?
ii. (15 points) Determine the optimal two-part tariff for each buyer. Compare the sales
and social welfare under the two-part tariffs with those in (i).
The optimal two-part tariff means fixing P=MC and charging each
buyer the fixed fee equal to his consumer surplus at P=1.
1 5 1 7 8
CS1 (P = 1) = · 3 · ( − 1) = 2.25; CS2 (P = 1) = · 4 · ( − 1) = .
2 2 2 3 3
So, the optimal two-part tariff is
8 59
P1 = 1, F1 = 2.25; P2 = 1, F2 = Π(ii) = SW(ii) =
3 12
Q(ii) = 7, Q1(i) = 1.5, Q2(i) = 2 ⇒ Q(i) = 3.5 < Q(ii) = 7
9 4 59
Π1(i) = 1.5 · 0.75 = 1.125; Π2(i) = 2 · 2/3 = 4/3. Π(i) =
+ =
8 3 24
1 3 3 9 1 2 2 59
CS1(i) (P = 1.75) = · · = ; CS2(i) (P = 5/3) = · 2 · = , CS(i) =
2 2 4 16 2 3 3 48
59 59 59 59
SW(i) = CS(i) + P i(i) = + = < = SW(ii)
48 24 16 12
In fact, the sales and social welfare under two-part tariff are equal to
the ones under perfect competition and thus are higher that under price-
discrimination.
iii. (5 points) Suppose firm cannot distinguish buyers, but still wants to use the two-part
tariff. What will be the common two-part tariff the firm will offer to both buyers?
If the firm decides to cater only to the second buyer, who pays a higher
fixed fee, its profit will be Π = 8/3. To keep both buyers, the firm has to
offer P = 1, F = 9/4 to both. It will then make Π = 2 · 9/4 = 9/2 > 8/3.
So, the optimal two-part tariff is P = 1, F = 9/4.
10