Escolar Documentos
Profissional Documentos
Cultura Documentos
A(p 0 , u 0 ) = {x p 0 x E (p 0 , u 0 )}
u(x ) = min v (p , p x )
n p R++
Lets find the direct utility function that generates this indirect
1
utility. Set y = 1, then v (p1 , p2 , 1) = (p1r + p2r ) r .
1
u(x1 , x2 ) = min p1 , p2 (p1r + p2r ) r s.t. p1 x1 + p2 x2 1 = 0
1
L(p1 , p2 , ) = (p1r + p2r ) r (p1 x1 + p2 x2 1)
Q(y ) = {x 0f (x ) = y }
MRTS is similar to MRS for consumers: the rate at which one input
can be substituted for another, while keeping output level constant.
f (x )
We will denote xi
as fi (x ).
xj0
r= xi0
, the ratio of xj to xi at x 0
x x
d ln ( xji ) d ln ( xji )
=
d ln MRTSij df df
d ln ( dx / dxj )
i
df df
d(xj /xi ) dxi / dxj
=
xj /xi d ( df / df )
dxi dxj
x2 1
MRTS12 (x1 , x2 ) = ( )
x1
x2
Set r = x1
.
d ln MRTS12 (x (r )) d ln r 1
=
d ln r d ln r
d ln r
= (1 ) =1
d ln r
Elasticity is a constant, hence CES.
j xj 1
MRTSij (xi , xj ) = ( )
i xi
j
As 0, MRTSij = i
( xxji )
This is the MRTS of the Cobb-Douglas production function:
y = x11 x22 ...xnn
1
CES: y = (x1 + x2 ) , 0 < 1
ij = 1/(1 ) for all i j.
As , ij 0, there is no substitutability between outputs
This is the Leontief production function:
y = min(x1 , ..., xn )
x1 x2
f( + )1
y1 + y2 y1 + y2
x1 x2
f( + )1
y1 + y2 y1 + y2
By linear homogeneity of f ,
f (x 1 + x 2 ) y 1 + y 2 = f (x 1 ) + f (x 2 )
f (t x 1 + (1 t)x 2 ) f (t x 1 ) + f (t x 2 ) = tf (x 1 ) + (1 t)f (x 2 )
fi (x )xi MPi (x )
i (x ) = =
f (x ) APi (x )
y = k(1 + x1 x2 )1
f1 (x )x1 x1 x2 f2 (x )x2 x1 x2
1 (x ) = = , ( x ) = =
f (x ) f (x )
2
1 + x1 x2 1 + x1 x2
( + )x1 x2
(x ) =
1 + x1 x2
Elasticities depend on x1 , x2 .
x1 x2 y
2 (y ) = = (1 )
1 + x1 x2 k
y
(y ) = ( + ) (1 )
k
Returns to scale decline monotonically as output increases.
At y = 0, u (y ) = + > 0. At y k, u (y ) 0.
If + > 1, returns to scale will go from increasing to decreasing as
y goes up.
The cost function, for strictly positive input prices w and feasible
output levels y f (Rn+ ), is defined as:
c(w , y ) = w x (w , y )
L(x1 , ..., xn , ) = w x (f (x ) y )
L f (x )
= wi
xi xi
f (x )
wi = for i = 1, ..., n
xi
f (x )/xi wi
MRTSi,j = =
f (x )/xj wj
First-order conditions:
w1 x1 1
MRTS12 = =( )
w2 x2
1
1 1
y = (x1 + x2 ) = x2 w21 (w11 + w21 )
Conditional input demands:
1
1
x1 = yw11 (w11 + w21 )
1
1
x2 = yw21 (w11 + w21 )
1
1
x1 = yw11 (w11 + w21 )
1
1 1
x2 = yw21 (w11 + w21 )
Cost function:
c(w 0 , y 0 )
= xi (w 0 , y 0 ) for i = 1, ..., n
wi
c(w1 , w2 , y )
x1 (w1 , w2 , y ) = Aw11 w2 y =
w1
c(w1 , w2 , y )
x2 (w1 , w2 , y ) = Aw1 w21 y =
w2
Ratio of conditional input demands:
x1 (w1 , w2 , y ) w2
=
x2 (w1 , w2 , y ) w1
(w ,y )
This implies that xiw i
0 for all i. (Recall: Hicksian
demand always decreases as own price increases).
c(w , y ) = y c(w , 1)
1
x (w , y ) = y x (w , 1)
1
So far, we have been assuming that all inputs can be changed when
calculating the appropriate cost.
This may not be true in all situations. In particular, in the short run,
some inputs may be fixed (e.g. if capital inputs, such as factories
or machines, take a long time to build).
We denote the long-run cost function as the one where all inputs
can be changed.
The short-run cost function is where some inputs are fixed.
z = (x , x )
sc(w , w , y ; x ) = w x (w , w , y ; x ) + w x
For a given level of output, long-run costs (where the firm can set
all input levels) cannot be greater than short-run costs (where the
firm cannot set all input levels).
This is because the feasible region for z of the short-run problem is
a subset of the feasible region of the long-run problem.
Any feasible z in the short-run problem is feasible in the long-run
problem, but not vice versa.
In general, if set A B, then for any function g :