Warwick University Department of Economics Prof.

Francesco Squintani

EC9D31 Advanced Microeconomics

Final Exam 2020-21 - Section A
Questions and Answers

Question 1. Suppose preferences take the form:

u(x1 ; x2 ) = minf2x1 ; x2 =2g:

(a) Derive the Marshallian demands xi (p; m), i = 1; 2. Are the goods Marshallian
complements or substitutes? (5 marks)
(b) Derive the indirect utility function v(p; m). Show that it is homogeneous of
degree zero in prices and income. (4 marks)
(c) Derive the expenditure function e(p; U ). Show that it is homogenous of degree 1
in prices. (4 marks)
(d) Derive the Hicksian demands hi (p; U ), i = 1; 2. Are the goods Hicksian comple-
ments or substitutes? (4 marks)
(e) Suppose that a third good x3 becomes available, such that preferences take now
the form
u(x1 ; x2 ; x3 ) = minf2x1 + x2 ; x3 =2g:

Derive Marshallian demands, xi (p; m), and Hicksian demands, hi (p; U ), i =

1; 2; 3 . [Hints. For what prices does the consumer simultaneously consume
goods 1 and 3? What happens for all other prices?] (8 marks)

Answers to Q1 We proceed in sequence as follows.

(a) The optimal choice occurs at 2x1 = x2 =2; and hence x2 = 4x1 ; so that the budget
constraint budget constraint p1 x1 + p2 x2 = p1 x1 + p2 (4x1 ) = y:
Solving out, x1 (p; y) = y=(p1 + 4p2 ); and hence x2 = 4y=(p1 + 4p2 ).

1
 (b) Substituting the Marshallian demands into the utility formula: v(p; y) = 2x1 (p; y) =
2y=(p1 + 4p2 ):
(c) By setting v = u and y = e in v( p,y) and solving for e(p,u) we get e(p; u) =
p1 u=2 + p2 2u = (p1 =2 + 2p2 )u:
de(p;u) de(p;u)
(d) Using Shepard Lemma dp1
= u=2 = h1 (p; U ) and dp2
= 2u = h2 (p; U ):
(e) At the optimum x1 + 2x2 = x3 =2 and the consumer consumes x1 = 0 if p1 > p2 =2
and x2 = 0 if p1 < p2 =2:

In the …rst case, 2x2 = x3 =2; and hence x3 = 4x2 ; so that the budget constraint
p1 x1 + p2 x2 + p3 x3 = p2 x2 + p3 (4x2 ) = y:
Solving out, x2 (p; y) = y=(p2 + 4p3 ); and hence x3 = 4y=(p2 + 4p3 ) and v(p; y) =
2x2 (p; y) = 2y=(p2 + 4p3 ):
Further, because u = 2x2 ; it follows that h2 = u=2; and hence that x3 = 2u and
e(p; u) = p2 u=2 + p3 2u = (p2 =2 + 2p3 )u:

In the second case, x1 = x3 =2; so that the budget constraint p1 x1 +p2 x2 +p3 x3 =
p1 x1 + p3 (2x1 ) = y:
Solving out, x1 (p; y) = y=(p1 + 2p3 ); and hence x2 = 2y=(p1 + 2p3 ) and v(p; y) =
x2 (p; y) = y=(p2 + 2p3 ):
Further, because u = x1 ; it follows that h1 = u, h3 = 2u and e(p; u) = p1 u +
p3 (2u) = (p1 + 2p3 )u:

Question 2. Consider a Cobb-Douglas Production function:

f (x) = x1 x2

where > 0, > 0 and make no assumptions on + .

(a) Set up the cost minimization problem and write up the Lagrangian. (5 marks)
(b) Derive the conditional factor demands h1 (w; y) and h2 (w; y). (5 marks)
(c) Find the 2 2 matrix of marginal price e¤ects. Con…rm the signs (and, where
appropriate, relative magnitudes) of these e¤ects. (5 marks)
(d) Find the cost function c(w; y). Con…rm its properties. (5 marks)
(e) Prove the following result: A technology exhibits CRS if and only if the produc-
tion function f (x) (if available) is homogeneous of degree 1. (5 marks)

2
Answers to Q2 We proceed in sequence as follows.

(a) The cost minimization problem is:

min w1 x1 + w2 x2 s.t. x1 x2 y:
x1 ;x2

The consequent Lagrangian is:

L = w 1 x1 + w 2 x2 (x1 x2 y)

(b) The conditional factor demands are:

+ + 1
h1 (w1 ; w2 ; y) = y +
w1 w2

and
+ + 1
h2 (w1 ; w2 ; y) = y + :
w1 w2
(c) The matrix of marginal price e¤ects is:
" # " #
@h1 @h1 1 1
@w1 @w2 w1 +
h1 w2 +
h1
@h2 @h2
= 1 1
:
@w1 @w2 w1 +
h2 w2 +
h2

(d) The cost function is:

1 w1 + w2 +
c(w; y) = w1 h1 (w; y) + w2 h2 (w; y) = y + ( + )

(e) Assume CRS: this implies that if z 2 Z then t z 2 Z, for all t 0. By de…nition,
z 2 Z means y f (x) and t z 2 Z means t y f (t x). By de…nition of
f (x) choose z, and hence x and y, so that y = f (x). We can then re-write the
condition above as: t f (x) f (t x). We need to prove that the equality holds.
Suppose it does not. Then there exists y 0 such that!t f (x) < y 0 < f (t x) Now
tx
y 0 < f (t x) implies by de…nition of Z that 2 Z and by CRS we get
y0
! !
1 tx x
0
2 Z, or 1 0
2 Z which means (1=t) y 0 f (x), or y 0 t f (x).
t y t
y

3
 This latter inequality contradicts t f (x) < y 0 .
The opposite implication is an immediate consequence of the de…nition of ho-
mogeneity of degree 1.

Question 3. There are two consumers A and B with the following utility functions and
endowments, with ! 1 ! 2 , 2 [0; 1] and 1:

uA = ln x1A + (1 ) ln x2A ; ! A = (0; ! 2 )

uB = minf x1B ; (1 )x2B g; ! B = (! 1 ; 0):

(a) Derive the Marshallian demands xi (p; m), i = A; B. (5 marks)

(b) Calculate the market clearing prices and the equilibrium allocations. (5 marks)
(c) Explain how the Walrasian equilibrium price of good 1 varies with and . (5
marks)
(d) Calculate the e¤ect of an increase in ! 1 or ! 2 on the equilibrium price of good
1. (5 marks)
(e) In general, an allocation (x1 ; x2 ; :::; xL ) in an exchange economy is said to be
Pareto-e¢ cient if there does not exist another feasible allocation (x01 ; x02 ; :::; x0L )
such that: (a) ul (x0l ) ul (xl ); for all l; and (b) ul (xl ) > ul (xl ); for some l.
Prove that a Walrasian equilibrium allocation (x1 ; x2 ; :::; xL ) is Pareto-e¢ cient.
(5 marks)

Answers to Q3 We proceed in sequence as follows.

(a) Let p be the price of good 1 and normalize p2 = 1:

Given price p; consumer A chooses xA so that

max ln x1A + (1 ) ln x2A s:t: px1A + x2A = ! 2 :

Hence,
max ln x1A + (! 2 ) ln(! 2 px1A );

4
 …rst-order conditions are:
(1 )
=p ;
x1A !2 px1A

solving out, x1A = ! 2 =p; substituting back, we obtain: x2A = ! 2 (1 ):

Given price p; consumer B chooses xB so that

max minf x1B ; (1 )x2B g s:t: px1B + x2B = p! 1 :

The consumer chooses x1B = (1 ) x2B ; solving this together with px1B + x2B =
p! 1 yields:
p (1 ) !1 !1
x1B = ; x2B = p
p (1 )+ p (1 )+
(b) Market clearing condition, therefore, is:

!2 p (1 ) !1
x1A + x1B = + = !1
p p (1 )+

Hence the equilibrium price is:

!2
p=
!1 ! 2 (1 )

and the equilibrium allocations are

(1 )
x1A = ! 1 !2 ; x2A = ! 2 (1 );
1
x1B = ! 2 ; x2B = ! 2 :

(c) The price p of good 1 is:

!2
p= ;
!1 ! 2 (1 )

di¤erentiating with respect to and , I obtain:

2
@ !2 !1!2
= >0
@ !1 ! 2 (1 ) ( !1 ! 2 (1 ))2

2
@ !2 ! 22
= < 0:
@ !1 ! 2 (1 ) ( !1 ! 2 (1 ))2
The equilibrium price of good 1 increases in and decreases in :

5
 (d) Di¤erentiating with respect to ! 1 and ! 2 , I obtain:

2
@ !2 !2
= < 0;
@! 1 !1 ! 2 (1 ) ( !2 + !1 + ! 2 )2

2
@ !2 !1
= > 0:
@! 2 !1 ! 2 (1 ) ( !2 + !1 + ! 2 )2
The equilibrium price of good 1 decreases in ! 1 and increases in ! 2 :
(e) Assume that the result is not true. There exists an allocation x such that
PI i
i=1 x !, ui (xi ) ui (xi; ) for all i and ui (xi ) > ui (xi; ) for some i.
Then, let’s …rst show that, for all i,

p xi p xi; : (1)

Assume that this is not true and there exists i such that p xi < p xi; . From
p xi; = p ! i we then get p xi < p ! i . This implies that there exists " > 0
such that if we denote eT the vector eT = (1; : : : ; 1), then p (xi + " e) < p ! i :
Monotonicity of preferences then implies that ui (xi +" e) > ui (xi ) which together
with the contradiction hypothesis gives: u(xi + " e) > ui (xi; ). This contradicts
xi; = xi (p ).
Since for some i we have ui (xi ) > ui (xi; ) then let’s show that, for the same i,

p xi > p xi; : (2)

Assume this is not the case. Then there exists a consumption bundle xi which
is a¤ordable for i: p xi p xi; = p ! i and yields a higher level of utility:
ui (xi ) > ui (xi; ). This is a contradiction of the hypothesis xi; = xi (p ).
PI i
Adding up Conditions (1) and (2) across consumers we obtain: i=1 p x >
PI P P
i=1 p x
i;
or Ii=1 p xi > Ii=1 p xi; = p !. This is a contradiction of the
feasibility of the allocation x.

Question 4. There are three individuals in society, f1; 2; 3g; three alternatives, fx; y; zg;
and the domain of preferences is unrestricted. Suppose that the social preference
relation, R, is given by pairwise majority voting (where voters break any indi¤erences
by voting for x …rst then y then z) if this results in a transitive social order. If this

6
 does not result in a transitive social order the social order is xP yP z. Let f denote
the social welfare function that this de…nes.

(a) Consider the following pro…les, where Pi is individual i’s strict preference relation:
Individual 1: xP1 yP1 z
Individual 2: yP2 zP2 x
Individual 3: zP3 xP3 y
What is the social order? (3 marks)
(b) What would be the social order if individual 1’s preferences in (a) were instead
yP1 zP1 x? or instead zP1 yP1 x? (5 marks)
(c) Prove that f satis…es the Pareto property, WP. (3 marks)
(d) Prove that f is non-dictatorial. (3 marks)
(e) Conclude that f does not satisfy IIA. (3 marks)
(f) Prove the following result: A social welfare rule is majoritarian if and only if it
is neutral, anonymous, and positively responsive. (8 marks)

Answers to Q4 We proceed in sequence as follows.

(a) The preferences xP1 yP1 z; yP2 zP2 x; zP3 xP3 y determine a Condorcet cycle, hence
the social order is xP yP z:
(b) With preferences yP1 zP1 x; yP2 zP2 x; zP3 xP3 y; the social order is yP zP x: With
preferences zP1 yP1 x; yP2 zP2 x; zP3 xP3 y; the social order is zP yP x
(c) The social choice function f satis…es Weak Pareto: if xPi y for all i; then x and
y cannot be part of a Condorcet cycle, and xP y: Thus, y 6= f (R):
(d) The social choice function f is not dictatorial: consider any agent i and pair of
alternatives x; y such that xPi y: Consider the pro…le of opponents’preferences
R i such that y is at the top of Rj and x is at the bottom, for all j 6= i: Then x
and y cannot be part of a Condorcet cycle, and yP x:
(e) The social choice function f cannot satisfy IIA, or else this would be a violation
of Arrow impossibility theorem.
(f) Suppose that there are only two alternatives: x is the status quo, and y is the
alternative. Each individual preference R(i) is indexed as q in f 1; 0; 1g, where 1

7
is a strict preference for x. The social welfare rule is a functional F (q(1); :::; q(N ))
in f 1; 0; 1g:
The social rule F is anonymous if for every permutation p, F (q(1); :::; q(N )) =
F (q(p(1)); :::; q(p(N ))):
The social rule F is neutral if F (q) = F ( q).
The rule F is positively responsive if q q 0 , q 6= q 0 and F (q 0 ) 0 imply that
F (q) = 1.
A social welfare rule F is majoritarian if:
. F (q) = 1 if and only if: n+ (q) = #fi : q(i) = 1g > n (q) = #fi : q(i) = 1g;
. F (q) = 1 if and only if n+ (q) < n (q);
. F (q) = 0 if and only if n+ (q) = n (q)).
May’s Theorem A social welfare rule is majoritarian if and only if it is neutral,
anonymous, and positively responsive.
Proof: Clearly, majority rule satis…es the 3 axioms.
By anonimity F (q) = G(n+ (q); n (q)). If n+ (q) = n (q), then n+ ( q) =
n ( q), and so, by neutrality, F (q) = G(n+ (q); n (q)) = G(n+ ( q); n ( q)) =
F ( q) = F (q) . This implies that F (q) = 0. If n+ (q) > n (q); pick q 0 with
q 0 < q and n+ (q 0 (q 0 ). Because F (q 0 ) = 0, by positive responsiveness, it follows
that F (q) = 1. When n+ (q) < n (q), it follows that n+ ( q) > n ( q), hence
F ( q) = 1 and by neutrality, F (q) = 1.