Envelope Theorem

Kevin Wainwright
Mar 22, 2004
1 Maximum Value Functions
A maximum (or minimum) value function is an objective function where the choice
variables have been assigned their optimal values. These optimal values of the choice
variables are, in turn, functions of the exogenous variables and parameters of the
problem. Once the optimal values of the choice variables have been substituted into
the original objective function, the function indirectly becomes a function of the
parameters (through the parameters inuence on the optmal values of the choice
variables). Thus the maximum value function is also referred to as the indirect
objective function.
What is the signicance of the indirect objective function? Consider that in any
optimization problem the direct objective function is maximized (or minimized) for a
given set of parameters. The indirect objective function gives all the maximum values
of the objective function as these prameters vary. Hence the indirect objective func-
tion is an envelope of the set of optimized objective functions generated by varying
the parameters of the model. For most students of economics the rst illustration
of this notion of an envelope arises in the comparison of short-run and long-run
cost curves. Students are typically taught that the long-run average cost curve is an
envelope of all the short-run average cost curves (what parameter is varying along
the envelope in this case?). A formal derivation of this concept is one of the exercises
we will be considering in the following sections.
To illustrate, consider the following maximization problem with two choice vari-
ables x and y, and one parameter, :
Maximize
U = f(x, y, ) (1)
The rst order necessary condition are
f
x
(x, y, ) = f
y
(x, y, ) = 0 (2)
if second-order conditions are met, these two equations implicitly dene the solutions
x = x

() y = x

() (3)
1
If we subtitute these solutions into the objective function, we obtain a new function
V () = f(x

(), y

(), ) (4)
where this function is the value of f when the values of x and y are those that
maximize f(x, y, ). Therefore, V () is the maximum value function (or indirect
objective function). If we dierentiate V with respect to
V

= f
x
x

+f
y
y

+f

(5)
However, from the rst order conditions we know f
x
= f
y
= 0. Therefore, the
rst two terms disappear and the result becomes
V

= f

(6)
This result says that, at the optimum, as varies, with x

and y

allowed to
adjust optimally gives the same result as if x

and y

were held constant! Note that

enters maximum value function (equation 4) in three places: one direct and two
indirect (through x

and y

). Equations 5 and 6 show that, at the optimimum, only

the direct eect of on the objective function matters. This is the essence of the
envelope theorem. The envelope theorem says only the direct eects of a change in
an exogenous variable need be considered, even though the exogenous variable may
enter the maximum value function indirectly as part of the solution to the endogenous
choice variables.
1.1 The Prot Function
Lets apply the above approach to an economic application, namely the prot function
of a competitive rm. Consider the case where a rm uses two inputs: capital, K,
and labour, L. The prot function is
= pf(K, L) wL rK (7)
where p is the output price and w and r are the wage rate and rental rate respec-
tively.
The rst order conditions are

L
= f
L
(K, L) w = 0

K
= f
K
(K, L) r = 0
(8)
which respectively dene the factor demand equations
L = L

(w, r, p)
K = K

(w, r, p)
(9)
2
substituting the solutions K

and L

into the objective function gives us

(w, r, p) = pf(K

, L

) wL

rK

(10)

(w, r, p) is the prot function (or indirect objective function). The prot func-
tion gives the maximum prot as a function of the exogenous variables w, r, and
p.
Now consider the eect of a change in w on the rms prots. If we dierentiate
the original prot function (equation 7) with respect to w, holding all other variables
constant and we get

w
= L (11)
However, this result does not take into account the prot maximizing rms abil-
ity to make a substitution of capital for labour and adjust the level of output in
accordance with prot maximizing behavior.
Since

(w, r, p) is the maximum value of prots for any values of w, r, and

p, changes in

from a change in w takes all captial for labour subsitutions into

account. To evaluate a change in the maximum prot function from a change in w,
we dierentiate

(w, r, p) with respect to w yielding

w
= [pf
L
w]
L

w
+ [pf
K
r]
K

w
L

(12)
From the rst order conditions, the two bracketed terms are equal to zero. There-
fore, the resulting equation becomes

w
= L

(w, r, p) (13)
This result says that, at the the prot maximizing position, a change in prots with
respect to a change in the wage is the same whether or not the factors are held constant
or allowed to vary as the factor price changes. In this case the derivative of the prot
function with respect to w is the negative of the factor demand function L

(w, r, p).
Following the above procedure, we can also show the additional comparative statics
results

(w, r, p)
r
= K

(r, w, p) (14)
and

(w, r, p)
p
= f(K

, L

) = q

(15)
The simple comparative static results derived from the prot function is known
as Hotellings Lemma. Hotellings Lemma is simply an application of the envelope
theorem.
3
1.2 The Envelope Theorem and Constrained Optimization
Now let us turn our attention to the case of constrained optimization. Again we will
have an objective function (U), two choice variables, (x and y) and one prarameter
() except now we introduce the following constraint:
g(x, y; ) = 0
The derivation of the envelope theorem for the models with one constraint is as
follows:
The problem then becomes
Maximize
U = f(x, y; ) (16)
subject to
g(x, y; ) = 0 (17)
The Lagrangian for this problem is
Z = f(x, y; ) +g(x, y; ) (18)
The rst order conditions are
Z
x
= f
x
+g
x
= 0
Z
y
= f
y
+g
y
= 0
Z

= g(x, y; ) = 0
(19)
Solving this system of equations gives us
x = x

() y = y

() =

() (20)
Substituting the solutions into the objective function, we get
U

= f(x

(), y

(), ) = V () (21)
where V () is the indirect objective function, or maximum value function. This
is the maximum value of y for any and x
i
s that satisfy the constraint.
How does V () change as changes? First, we dierentiate V with respect to
V

= f
x
x

+f
y
y

+f

(22)
In this case,equation 22 will not simplify to
V

= f

since f
x
6= 0 and f
y
6= 0.
However, if we substitute the solutions to x and y into the constraint (producing an
identity)
g(x

(), y

(), ) 0 (23)
4
and dierentiating with respect to yields
g
x
x

+g
x
x

+g

0 (24)
If we multiply equation 24 by and combine the result with equation 22 and
rearranging terms, we get
V

= (f
x
+g
x
)
x

+ (f
y
+g
y
)
y

+f

+g

= Z

(25)
Where Z

is the partial deviative of the Lagrangian function with respect to ,

holding all other variable constant. In this case, the Langrangian functions serves as
the objective function in deriving the indirect objective function.
While the results in equation 25 nicely parallel the unconstrained case, it is impor-
tant to note that some of the comparative static results depend critically on whether
the parameters enter only the objective function or whether they enter only the con-
straints, or enter both. If a parameter enters only in the objective function then the
comparative static results are the same as for unconstrained case. However, if the
parameter enters the constraint, the relation
V

will no longer hold.

1.3 Interpretation of the Lagrange Multiplier
In the consumer choice problem in chapter 12 we derived the result that the Lagrange
multiplier, , represented the change in the value of the Lagrange function when
the consumers budget changed. We loosely interpreted as the marginal utility of
income. Now let us derive a more general interpretation of the Lagrange multiplier
with the assistance of the envelope theorem.Consider the problem
Maximize
U = f(x, y) (26)
Subject to
c g(x, y) = 0 (27)
where c is a constant. The Lagrangian for this problem is
Z = f(x, y) +(c g(x, y)) (28)
The rst order equations are
Z
x
= f
x
(x, y) g
x
(x, y) = 0
Z
y
= f
y
(x, y) g
y
(x, y) = 0
Z

= c g(x, y) = 0
(29)
5
From the rs twot equations in (29), we get
=
f
x
g
x
=
f
y
g
y
(30)
which gives us the condition that the slope of the level curve of the objective function
must equal the slope of the constraint at the optimum.
Equations (29) implicitly dene the solutions
x = x

(c) y = y(c) =

(c) (31)
substituting (31) back into the Lagrangian yields the mamximum value function
V (c) = Z

(c) = f (x

(c), y

(c)) +

(c) (c g(x

1
(c), y

(c))) (32)
dierentiating with respect to c yields
Z

c
= f
x
x

c
+f
y
y

c
+(c g(x

(c), y

(c)))

(c)g
x
x

(c)g
y
y

c
+

(c)
c
c
(33)
by rearranging we get
Z

c
= (f
x

g
x
)
x

c
+ (f
y

g
y
)
y

c
+ (c g(x

, y

))

c
+

(34)
Note that the three terms in brackets are nothing more than the rst order equa-
tions and, at the optimal values of x, y and , these terms are all equal to zero.
Therefore this expression simplies to
V (c)
c
=
Z

c
=

(35)
Therefore equals the rate of change of the maximum value of the objective function
when c changes ( is sometimes referred to as the shadow price of c).Note that, in
this case, c enters the problem only through the constraint; it is not an argument of
the original objective function.
2 Duality and the Envelope Theorem
A consumers expenditure function and his indirect utility function are the minimum
and maximum value functions for dual problems. An expenditure function species
the minimum expenditure required to obtain a xed level of utility given the utility
function and the prices of consumption goods. An indirect utility function speci-
es the maximum utility that can be obtained given prices, income and the utility
function.
6
Let U(x, y) be a utility function in x and y are consumption goods. The consumer
has a budget, B, and faces market prices P
x
and P
y
for goods x and y respectively.
Setting up the Lagrangian:
Z = U(x, y) +(B P
x
x P
y
y) (36)
The rst order conditions are
Z
x
= U
x
P
x
= 0
Z
y
= U
y
P
y
= 0
Z

= B P
x
X P
y
Y = 0
(37)
This system of equations implicity denes a solution for x
M
, y
M
and
M
as a
function of the exogenous variables B, P
x
, P
y
.
x
M
= x
M
(P
x
, P
y
, B)
y
M
= y
M
(P
x
, P
y
, B)

M
=
M
(P
x
, P
y
, B, )
(38)
The solutions to x
M
and y
M
are the consumers ordinary demand functions, some-
times called the Marshallian demand functions.
1
Substituting the solutions to x

and y

into the utility function yields

U

= U

(x
M
(B, P
x
, P
y
), y
M
(B, P
x
, P
y
)) = V (B, P
x
, P
y
) (39)
Where V is the maximum value function, or indirect utility function.
Now consider the alternative, or dual, problem for the consumer; minimize total
expenditure on x and y while maintaining a given level of utility, U

. The Langranian
for this problem is
Z = P
x
x +P
y
y +(U

U(x, y)) (40)

The rst order conditions are
Z
x
= P
x
U
x
= 0
Z
y
= P
y
U
y
= 0
Z

= U

U(x, y; ) = 0
(41)
This system of equations implicitly dene the solutions to x
h
, y
h
and
h
x
h
= x
h
(U

, P
x
, P
y
)
y
h
= y
h
(U

, P
x
, P
y
)

h
=
h
(U

, P
x
, P
y
)
(42)
1
Named after the famous economist Alfred Marshall, known to most economic students as an-
other dead guy.
7
x
h
and y
h
are the compensated, or real income held constant demand func-
tions. They are commonly referred to as Hicksion demand functions, hence the h
superscript.
2
If we compare the rst two equations from the rst order conditions in both utility
maximization problem and expenditure minimization problem (Z
x
, Z
y
), we see that
both sets can be combined (eliminating ) to give us
Px
Py
=
Ux
Uy
(= MRS) (43)
This is the tangency condition in which the consumer chooses the optimal bundle
where the slope of the indierence curve equals the slope of the budget constraint.
The tangency condition is identical for both problems. If the target level of utility
in the minimization problem is set equal to the value of the utility obtained in the
solution to the maximization problem, namely U

, we obtain the following

x
M
(B, P
x
, P
y
) = x
h
(U

, P
x
, P
y
)
y
M
(B, P
x
, P
y
) = y
h
(U

, P
x
, P
y
)
(44)
or the solution to both the maximization problelm and the minimization problem
produce identical values for x and y. However, the solutions are functions of dierent
exogenous variables so any comparative statics exercises will produce dierent results.
Substituting x
h
and y
h
into the objective function of the minimization problem
yields
P
x
x
h
(P
x
, P
y
, U

) +P
y
y
h
(P
x
, P
y
, U

) = E(P
x
, P
y
, U

) (45)
where E is the minimum value function or expenditure function. The duality
relationship in this case is
E(P
x
, P
y
, U

, ) = B (46)
where B is the exogenous budget from the maximization problem.
Finally, it can be shown from the rst order conditions of the two problems that

M
=
1

h
(47)
2.1 Roys Identity
One application of the envelope theorem is the derivation of Roys identity. Roys
identity states that the individual consumers marshallian demand function is equal
to the ratio of partial derviatives of the maximum value function. Substituting the
optimal values of x
M
, y
M
and
M
into the Lagrangian gives us
V (B, P
x
, P
y
) = U(x
M
, y
M
) +
M
(B P
x
x
M
P
y
y
M
) (48)
2
Yet another famous, but dead economist, Sir John Hicks.
8
First dierentiate with respect to P
x
V
P
x
= (U
x

M
P
x
)
x
M
P
x
+(U
y

M
P
y
)
y
M
P
x
+(BP
x
x
M
P
y
y
M
)

M
P
x

M
x
M
(49)
V
P
x
= (0)
x
M
P
x
+ (0)
y
M
P
x
+ (0)

M
P
x

M
x
M
=
M
x
M
(50)
Next, dierentiate the value function with respect to B
V
B
= (U
x

M
P
x
)
x
M
B
+(U
y

M
P
y
)
y
M
B
+B P
x
x
M
P
y
y
M
)

M
B
+
M
(51)
V
B
= (0)
x
M
B
+ (0)
y
M
B
+ (0)

M
B
+
M
=
M
(52)
Finally, taking the ratio of the two partial derivatives
V
P
x
V
B
=

M
x
M

M
= x
M
(53)
which is Roys identity.
2.2 Shephards Lemma
Earlier in the chapter an application of the envelope theorem was the derivation
of Hotellings Lemma, which states that the partial derivatives of the maximum
value of the prot function yields the rms factory demand functions and the supply
functions. A similar approach applied to the expenditure function yields Shepards
Lemma.
Consider the consumers minimization problem. The Lagrangian is
Z = P
x
x +P
y
y +(U

U(x, y)) (54)

From the rst order conditions, the solutions are implicitly dened
x
h
= x
h
(P
x
, P
y
, U

)
y
h
= y
h
(P
x
, P
y
, U

h
=
h
(P
x
, P
y
, U

)
(55)
Substituting these solutions into the Lagrangian yields the minimum value func-
tion
V (P
x
, P
y
, U

) = P
x
x
h
+P
y
y
h
+
h
(U

U(x
h
, y
h
)) (56)
9
The partial derivatives of the value function with respect to P
x
and P
y
are the
consumers conditional, or Hicksian, demands:
V
P
x
= (P
x

h
U
x
)
x
h
P
x
+ (P
y

h
U
y
)
y
h
P
x
+ (U

U(x
h
, y
h
))

h
P
x
+x
h
V
P
x
= (0)
x
h
P
x
+ (0)
y
h
P
x
+ (0)

h
P
x
+x
h
= x
h
(57)
and
V
P
y
= (P
x

h
U
x
)
x
h
P
y
+ (P
y

h
U
y
)
y
h
P
y
+ (U

U(x
h
, y
h
))

h
P
y
+y
h
V
Py
= (0)
x
h
Py
+ (0)
y
h
Py
+ (0)

h
Py
+y
h
= y
h
(58)
Dierentiating V with respect to the constraint U

yields
h
, the marginal cost of
the constraint
V
U

= (P
x

h
U
x
)
x
h
U

+ (P
y

h
U
y
)
y
h
P
y
+ (U

U(x
h
, y
h
))

h
U

+
h
V
U

= (0)
x
h
U

+ (0)
y
h
U

+ (0)

h
U

+y
h
=
h
Together, these three partial derivatives are Shepards Lemma.
2.3 Example of duality for the consumer choice problem
2.3.1 Utility Maximization
Consider a consumer with the utility function U = xy, who faces a budget constraint
of B = P
x
xP
y
y, where all variables are dened as before.
The choice problem is
Maximize
U = xy (59)
Subject to
B = P
x
xP
y
y (60)
The Lagrangian for this problem is
Z = xy +(B P
x
xP
y
y) (61)
The rst order conditions are
Z
x
= y P
x
= 0
Z
y
= x P
y
= 0
Z

= B P
x
x P
y
y = 0
(62)
Solving the rst order conditions yield the following solutions
x
M
=
B
2P
x
y
M
=
B
2P
y
=
B
2P
x
P
y
(63)
10
where x
M
and y
M
are the consumers Marshallian demand functions. Checking
second order conditions, the bordered Hessian is

0 1 P
x
1 0 P
y
P
x
P
y
0

= 2P
x
P
y
> 0 (64)
Therefore the solution does represent a maximum . Substituting x
M
and y
M
into
the utility function yields the indirect utility function
V (P
x
, P
y
, B) =

B
2P
x

B
2P
y

=
B
2
4P
x
P
y
(65)
If we denote the maximum utility by U
0
and re-arrange the indirect utility function
to isolate B
B
2
4P
x
P
y
= U
0
(66)
B = (4P
x
P
y
U
0
)
1
2
= 2P
1
2
x
P
1
2
y
U
1
2
0
= E(P
x
, P
y
, U
0
) (67)
We have the expenditure function
Roys Identity Lets verify Roys identity which states
x
M
=
V
P
x
V
B
(68)
Taking the partial derivative of V
V
P
x
=
B
2
4P
2
x
P
y
(69)
and
V
B
=
B
P
x
P
y
(70)
Taking the negative of the ratio of these two partials

V
P
x
V
B
=

B
2
4P
2
x
Py

B
P
x
P
y
=
B
2P
x
= x
M
(71)
Thus we nd that Roys Identity does hold.
11
2.3.2 The dual and Shepards Lemma
Now consider the dual problem of cost minimization given a xed level of utility.
Letting U
0
denote the target level of utility, the problem is
Minimize
P
x
x +P
y
y (72)
Subject to
U
0
= xy (73)
The Lagrangian for the problem is
Z = P
x
x +P
y
y +(U
0
xy) (74)
The rst order conditions are
Z
x
= P
x
y = 0
Z
y
= P
y
x = 0
Z

= U
0
xy = 0
(75)
Solving the system of equations for x, y and
x
h
=

P
y
U
0
P
x
1
2
y
h
=

PxU
0
Py
1
2

h
=

PxPy
U
0
1
2
(76)
where x
h
and y
h
are the consumers compensated (Hicksian) demand functions.
Checking the second order conditions for a minimum

0 y
0 x
y x 0

= 2xy < 0 (77)

Thus the sucient conditions for a minimum are satised.
Substituting x
h
and y
h
into the orginal objective function gives us the minimum
value function, or expenditure function
P
x
x
h
+P
y
y
h
= P
x

P
y
U
0
Px
1
2
+P
y

PxU
0
Py
1
2
= (P
x
P
y
U
0
)
1
2
+ (P
x
P
y
U
0
)
1
2
= 2P
1
2
x
P
1
2
y
U
1
2
0
(78)
Note that the expenditure function derived here is identical to the expenditure
function obtained by re-arranging the indirect utility function from the maximization
problem.
12
Shepards Lemma We can now test Shepards Lemma by dierentiating the ex-
penditure function directly.
First, we derive the conditional demand functions
E(P
x
, P
y
, U
0
)
P
x
=

P
x

2P
1
2
x
P
1
2
y
U
1
2
0

=
P
1
2
y
U
1
2
0
P
1
2
x
= x
h
(79)
and
E(P
x
, P
y
, U
0
)
P
y
=

P
y

2P
1
2
x
P
1
2
y
U
1
2
0

=
P
1
2
y
U
1
2
0
P
1
2
y
= y
h
(80)
Next, we can nd the marginal cost of utility (the Lagrange multiplier)
E(P
x
, P
y
, U
0
)
U
0
=

U
0

2P
1
2
x
P
1
2
y
U
1
2
0

=
P
1
2
x
P
1
2
y
U
1
2
0
=
h
(81)
Thus, Shepards Lemma holds in this example.
3 Income and Substitution Eects: The Slutsky
Equation
3.1 The Traditional Approach
Consider a representative consumer who chooses only two goods: x and y. The price
of both goods are determined in the market and are therefore exogenous. As well,
the consumers budget is also exogenously determined. The consumer choice problem
then is
Maximize
U(x, y) (82)
Subject to
B = P
x
X +P
y
Y (83)
The Langrangian function for this optimization problem is
Z = U(x, y) +(B P
x
x +P
y
y) (84)
The rst order conditions yield the following set of simultaneous equations:
Z

= B P
x
x P
y
y = 0
Z
x
= U
x
P
x
= 0
Z
y
= U
y
P
y
= 0
(85)
13
Solving this system will allow us to express the optimal values of the endogenous
variables as implicit functions of the exogenous variables:

(P
x
, P
y
, B)
x

= x

(P
x
, P
y
, B)
y

= y

(P
x
, P
y
, B)
If the bordered Hessian in the present problem is positive

0 P
x
P
y
P
x
U
xx
U
xy
P
y
U
yx
U
yy

= 2P
x
P
y
U
xy
P
2
y
U
xx
P
2
x
U
yy
> 0 (86)
then the value of U will be a maximum.
.
By substituting the optimal values x

, y

and

into the rst order equations, we

convert these equations into equilibrium identities:
B P
y
x

P
y
y

0
U
x
(x

, y

P
x
0
U
y
(x

, y

P
y
0
By taking the total dierential of each identity in turn, and noting that U
xy
= U
yx
(Youngs Theorem), we then arrive at the linear system
Pxdx

Pydy = x

dPx +y

dPy dB
Pxd

+U
xx
dx

+U
xy
dy

dPx
Pyd

+U
yx
dx

+U
yy
dy

dPy
(87)
Writing these equations in matrix form

0 P
x
P
y
P
x
U
xx
U
xy
P
y
U
yx
U
yy

dx

dy

dP
x
+y

dP
y
dB

dP
x

dP
y

To study the eect of a change in the budget, let the other exogenous dierentials
equal zero (dP
x
= dP
y
= 0, dB 6= 0). Then dividing through by dB, and applying the
implicit function theorem, we have

0 P
x
P
y
P
x
U
xx
U
xy
P
y
U
yx
U
yy

/B
dx

/B
dy

/B

1
0
0

(88)
The coecient matrix of this system is the Jacobian matrix, which has the same
value as the bordered Hessian

which is positive if the second order conditions are

met. By using Cramers rule we can solve for the following comparative static
x

B
=
1

0 1 P
y
P
x
0 U
xy
P
y
0 U
yy

=
1

P
x
U
xy
P
y
U
yy

H
12

0 (89)
14
As before, in the absence of additional information about the relative magnitudes
of P
x
, P
y
and the cross partials, U
ij
, we are unable to ascertain the sign of this
comparative-static derivative. This means that the optimal x

may increase in the

budget, B, depending on whether it is a normal or inferior good (ambiguous income
eect)
Next, we may analyze the eect of a change in P
x
. Letting dP
y
= dB = 0 but
keeping dP
x
6= 0 and dividing Equation 87 by dP
x
we obtain

0 P
x
P
y
P
x
U
xx
U
xy
P
y
U
yx
U
yy

/P
x
x

/P
x
y

/P
x

(90)
From this, the following comparative static emerges:
x

P
x
=
1
|H|

0 x

P
y
P
x

U
xy
P
y
0 U
yy

=
x

|
H
|

P
x
U
xy
P
y
U
yy

|
H
|

0 P
y
P
y
U
yy

= (x

)
|

H
12|
|
H
|
+

|

H
22|
|
H
|
(91)
Note that there are two componants in (
x

P
x
). By comparing the rst term to our
previous comparative static (
x

B
), we see that
(x

H
12

= (x

0 (92)
which can be interpreted as the income eect of a price change. The second term
is the income compensated version of x

/P
x
, or the substitution eect of a price
change, which is unambiguously negative:

P
x

compensated
=

0 P
y
P
y
U
yy

H
22

(P
2
y
) < 0 (93)
Hence, we can express Equation 91 in the form
x

| {z }
Income Eect
+

x

P
x

compensated
| {z }
Substitution Eect
(94)
This result, which decomposes the comparative static derivative (x

/P
x
) into
two componants, an income eect and a substitution eect, is the two-good version
of the Slutsky Equation.
15
3.2 Duality and the Alternative Slutsky
From the envelope theorem, we can derive the Slutsky decomposition in a more
succinct manner. Consider rst that from the utility maximum problem we derived
solutions for x and y
x
M
= x
M
(P
x
, P
y
, B)
y
M
= y
M
(P
x
, P
y
, B)
(95)
which were the marshallian demand functions. Substituting these solutions into the
utility function yielded the indirect utility function (or maximum value function)
U

= U(x
M
(P
x
, P
y
, B), y
M
(P
x
, P
y
, B)) = U

(P
x
, P
y
, B) (96)
which could be rewritten to isolate B and giving us the expenditure function
B

= B(P
x
, P
y,
U

) (97)
Second, from the budget minimization problem we derived the Hicksian, or com-
pensated, demand function
x

= x
h
(P
x
, P
y,
U

) (98)
which, by Shephards lemma, is equivalent to the partial derivative of the expenditure
function with respect to P
x
:
B(P
x
, P
y,
U

)
P
x
= x
c
(P
x
, P
y,
U

) (99)
Thus we know that if the maximum value of utility obtained from
Max U(x, y) +(B P
x
x P
y
y)
is the same value as the exogenous level of utility found in the constrained minimiza-
tion problem
Min P
x
x +P
y
y +(U
0
U(x, y)) (100)
the values of x and y that satisfy the rst order conditions of both problems will
be identical, or
x
c
(P
x
, P
y,
U
0
) = x
m
(P
x
, P
y,
B) (101)
at the optimum.If we subsitiute the expenditure function into x
M
in place of the
budget, B, we get
x
c
(P
x
, P
y,
U
0
) = x
M
(P
x
, P
y,
B

(P
x
, P
y,
U
0
)) (102)
Dierentiate both sides of equation 102 with respect to P
x
x
c
(Px,Py,U
0
)
P
x
=
x
M
(Px,Py,B

(Px,Py,U
0
))
P
x
+
x
M
(Px,Py,B

(Px,Py,U
0
))
B
B(Px,Py,U
0
)
P
x
(103)
16
But we know from Shephards lemma that
B(P
x
, P
y,
U
0
)
P
x
= x
c
(104)
substituting equation 104 in to equation 103 we get
x
c
P
x
=
x
M
P
x
+x
c
x
M
B
(105)
Subtract (x
c x
M
B
) from both sides gives us
x
M
P
x
= x
c
x
M
B
| {z }
Income eect
+
x
c
P
x
|{z}
Substitution eect
(106)
If we compare equation (106) to equation (94) we see that we have arrived at the
identical result. The method of deriving the slutsky decomposition through the ap-
plication of duality and the envelope theorem is sometimes referred to as the instant
slutsky.
3.2.1 Problems:
1. A consumer has the following utility function: U(x, y) = x(y+1), where x and y
are quantities of two consumption goods whose prices are p
x
and p
y
respectively.
The consumer also has a budget of B. Therefore the consumers maximization
problem is
x(y + 1) +(B p
x
x p
y
y)
(a) From the rst order conditions nd expressions for the demand functions.
What kind of good is y? In particular what happens when p
y
> B/2?
(b) Verify that this is a maximum by checking the second order conditions.
By substituting x

and y

into the utility function nd an expressions for

the indirect utility function
U

= U(p
x
, p
y
, B)
and derive an expression for the expenditure function
B

= B(p
x
, p
y
, U

)
(c) This problem could be recast as the following dual problem
Minimize p
x
x +p
y
y subject to U

= x(y + 1)
Find the values of x and y that solve this minimization problem and show
that the values of x and y are equal to the partial derivatives of the ex-
penditure function, B/p
x
and B/p
y
respectively.
17