Rational Choice

CHOICE
1. Scarcity (income constraint)
2. Tastes (indifference map/utility
function)
ECONOMIC RATIONALITY
◆ The principal behavioral postulate is
that a decision-maker chooses its
most preferred alternative from those
available to it.
◆ The available choices constitute the
choice set.
◆ How is the most preferred bundle in
the choice set located/found?
RATIONAL CONSTRAINED CHOICE
x2

More preferred
bundles

Affordable
bundles
x1
RATIONAL CONSTRAINED CHOICE
x2

x 2*

x 1* x1
RATIONAL CONSTRAINED CHOICE
x2 (x1*,x2*) is the most
preferred affordable
bundle.

x 2* E

x 1* x1
RATIONAL CONSTRAINED CHOICE
At Equilibrium E

MRS=∆x2/∆x1 = p1/p2

Slope of the Slope of the budget

indifference curve constraint

Individual’s willingness Society’s willingness to

to trade trade
RATIONAL CONSTRAINED CHOICE

◆ The most preferred affordable bundle

is called the consumer’s ORDINARY
DEMAND at the given prices and
income.
◆ Ordinary demands will be denoted by
x1*(p1,p2,m) and x2*(p1,p2,m).
RATIONAL CONSTRAINED CHOICE
x2
The slope of the
indifference curve at
(x1*,x2*) equals the
slope of the budget
x 2* constraint.

x 1* x1
RATIONAL CONSTRAINED CHOICE
(x1*,x2*) satisfies two conditions:
◆ (i) the budget is exhausted, i.e.
p1x1* + p2x2* = m; and
◆ (ii) the slope of the budget constraint,
(-) p1/p2, and the slope of the
indifference curve containing (x1*,x2*)
are equal at (x1*,x2*).
 COMPUTING DEMAND
◆ How can this information be used
to locate (x1*,x2*) for given p1, p2 and
m?
◆ Two ways to do this
3. Use Lagrange multiplier method
4. Find MRS and substitute into the
Budget Constraint
 COMPUTING DEMAND
Lagrange Multiplier Method
Suppose that the consumer has Cobb-
Douglas preferences
U ( x1 , x2 ) = x1a x12− a

and a budget constraint given by

p1 x1 + p2 x2 = m
 COMPUTING DEMAND
Lagrange Multiplier Method
Aim
a ( 1− a )
max x1 x 2 subject to p1 x1 + p2 x 2 = m

Set up the Lagrangian

L = x1a x 2( 1 − a ) + λ ( m − p1 x1 − p2 x 2 )
( x1 , x 2 , λ )
 COMPUTING DEMAND
Lagrange Multiplier Method
Differentiate

∂L
= ax1a −1 x2(1−a ) − λp1 = 0 (1)
∂x1
∂L
= (1 − a ) x1a x2( −a ) − λp2 = 0 (2)
∂x2
∂L
= m − p1 x1 − p2 x2 = 0 (3)
∂λ
 COMPUTING DEMAND
Lagrange Multiplier Method
From (1) and (2)

ax1( a-1 ) x 2( 1 −a ) ( 1 − a ) x1a x 2−a

λ= =
p1 p2

Then re-arranging
p1 ax2
=
p2 (1 − a ) x1
 COMPUTING DEMAND
Lagrange Multiplier Method
Rearrange
ap2 x 2
p1 x1 =
(1 − a )
Remember

p1 x1 + p2 x 2 = m
 COMPUTING DEMAND
Lagrange Multiplier Method
Substitute
ap2 x 2
p1 x1 =
(1 − a )

p1 x1 + p2 x 2 = m
 COMPUTING DEMAND
Lagrange Multiplier Method
Solve x1* and x2*

* am
x1 =
and
p1

*
x2 =
( 1 − a )m
p2
 COMPUTING DEMAND
Method 2

Suppose that the consumer has

Cobb-Douglas preferences.

U( x1 , x 2 ) = x1axb2
 COMPUTING DEMAND
Method 2
◆ Suppose that the consumer has
Cobb-Douglas preferences.
a b
U( x1 , x 2 ) = x1 x 2
∂U
MU1 = = ax1a − 1xb2
∂ x1

∂U
MU2 = = bx1axb2 − 1
∂ x2
 COMPUTING DEMAND
Method 2
◆ So the MRS is

a −1 b
dx 2 ∂ U/∂ x1 ax1 x 2 ax 2
MRS = =− =− =− .
dx1 ∂ U/∂ x 2 bx1axb2 −1 bx1
 COMPUTING DEMAND
Method 2
◆ So the MRS is

a −1 b
dx 2 ∂ U/∂ x1 ax1 x 2 ax 2
MRS = =− =− =− .
dx1 ∂ U/∂ x 2 bx1axb2 −1 bx1

◆ At (x1*,x2*), MRS = -p1/p2 , i.e. the

slope of the budget constraint.
 COMPUTING DEMAND
Method 2
◆ So the MRS is

a −1 b
dx 2 ∂ U/∂ x1 ax1 x 2 ax 2
MRS = =− =− =− .
dx1 ∂ U/∂ x 2 bx1axb2 −1 bx1

◆ At (x1*,x2*), MRS = -p1/p2 so

ax*2 p1 * bp1 *
− =− ⇒ x2 = x1 . (A)
bx*1 p2 ap 2
 COMPUTING DEMAND
Method 2
◆ (x1*,x2*) also exhausts the budget so
* *
p1x1 + p 2x 2 = m. (B)
 COMPUTING DEMAND
Method 2
◆ So now we know that
* bp1 *
x2 = x1 (A)
ap 2
* *
p1x1 + p 2x 2 = m. (B)
 COMPUTING DEMAND
Method 2
So now we know that
* bp1 *
x2 = x1 (A)
ap 2
Substitute
* *
p1x1 + p 2x 2 = m. (B)
 COMPUTING DEMAND
Method 2
So now we know that
* bp1 *
x2 = x1 (A)
ap 2
Substitute
* *
p1x1 + p 2x 2 = m. (B)
and get
* bp1 *
p1x1 + p 2 x1 = m.
ap 2
This simplifies to ….
COMPUTING DEMAND
Method 2
am
x*1 = .
( a + b )p1
 COMPUTING DEMAND
Method 2
* am
x1 =
(a + b ) p1
Substituting for x1* in
p1x*1 + p 2x*2 = m
then gives
bm
x 2* =
( a + b ) p2
 COMPUTING DEMAND
Method 2
So we have discovered that the most
preferred affordable bundle for a
consumer
with Cobb-Douglas preferences
a b
U( x , x ) = x x
1 2 1 2

is
( x1* , x 2* ) = ( am
,
bm
(a + b ) p1 (a + b) p2
)
 COMPUTING DEMAND
Method 2: Cobb-Douglas
x2 a b
U( x1 , x 2 ) = x1 x 2

*
x2 =
bm
( a + b )p 2

am x1
x*1 =
( a + b )p1
 Rational Constrained Choice
◆ But what if x1* = 0 or x2* = 0?
◆ If either x1* = 0 or x2* = 0 then the
ordinary demand (x1*,x2*) is at a
corner solution to the problem of
maximizing utility subject to a
budget constraint.
Examples of Corner Solutions:
Perfect Substitutes
x2
MRS = -1

x1
Examples of Corner Solutions:
Perfect Substitutes
x2
MRS = -1

Slope = -p1/p2 with p1 > p2.

x1
Examples of Corner Solutions:
Perfect Substitutes
x2
MRS = -1

Slope = -p1/p2 with p1 > p2.

x1
 Examples of Corner Solutions:
Perfect Substitutes
x2
MRS = -1 (This is the indifference
m curve)
x =
*
2
p2

Slope = -p1/p2 with p1 > p2.

x*1 = 0 x1
 Examples of Corner Solutions:
Perfect Substitutes
x2 ANOTHER
MRS = -1 EXAMPLE

Slope = -p1/p2 with p1 < p2.

x*2 = 0
x =
*
1
m x1
p1
 Examples of Corner Solutions:
Perfect Substitutes
So when U(x1,x2) = x1 + x2, the most
preferred affordable bundle is (x1*,x2*)
where m 
( x1 , x2 ) =  ,0 
* *
if p1 < p2
 p1 
or
 m
( x1 , x2 ) =  0, 
* *
if p1 > p2.
 p2 
Examples of Corner Solutions:
Perfect Substitutes
x2
m MRS = -1
p2 Slope = -p1/p2 with p1 = p2.

The budget
constraint and
the utility curve
lie on each other

m x1
p1
Examples of Corner Solutions:
Perfect Substitutes
x2
y All the bundles in the
p2 constraint are equally the
most preferred affordable
when p1 = p2.

y x1
p1
Examples of ‘Kinky’ Solutions:
Perfect Complements
X2 (gin) U(x1,x2) = min(ax1,x2)

x2 = ax1 (a = .5)

X1 (tonic)
Examples of ‘Kinky’ Solutions:
Perfect Complements
x2 U(x1,x2) = min(ax1,x2)

x2 = ax1
MRS = 0

x1
Examples of ‘Kinky’ Solutions:
Perfect Complements
x2 U(x1,x2) = min(ax1,x2)

MRS = - ∞

x2 = ax1
MRS = 0

x1
Examples of ‘Kinky’ Solutions:
Perfect Complements
x2 U(x1,x2) = min(ax1,x2)

MRS = - ∞
MRS is undefined
x2 = ax1
MRS = 0

x1
Examples of ‘Kinky’ Solutions:
Perfect Complements
x2 U(x1,x2) = min(ax1,x2)

x2 = ax1

x1
Examples of ‘Kinky’ Solutions:
Perfect Complements
x2 U(x1,x2) = min(ax1,x2)
Which is the most
preferred affordable bundle?

x2 = ax1

x1
Examples of ‘Kinky’ Solutions:
Perfect Complements
x2 U(x1,x2) = min(ax1,x2)
The most preferred
affordable bundle

x2 = ax1

x1
Examples of ‘Kinky’ Solutions:
Perfect Complements
x2 U(x1,x2) = min(ax1,x2)

x2 = ax1
x 2*

x 1* x1
Examples of ‘Kinky’ Solutions:
Perfect Complements
x2 U(x1,x2) = min(ax1,x2)
and p1x1* + p2x2* = m

x2 = ax1
x 2*

x 1* x1
Examples of ‘Kinky’ Solutions:
Perfect Complements
x2 U(x1,x2) = min(ax1,x2)
(a) p1x1* + p2x2* = m
(b) x2* = ax1*
x2 = ax1
x 2*

x 1* x1
 Examples of ‘Kinky’ Solutions:
Perfect Complements
(a) p1x1* + p2x2* = m; (b) x2* = ax1*
 Examples of ‘Kinky’ Solutions:
Perfect Complements
(a) p1x1* + p2x2* = m; (b) x2* = ax1*.
Substitution from (b) for x2* in
(a) gives p1x1* + p2ax1* = m
 Examples of ‘Kinky’ Solutions:
Perfect Complements
(a) p1x1* + p2x2* = m; (b) x2* = ax1*.
Substitution from (b) for x2* in
(a) gives p1x1* + p2ax1* = m
which gives x * = m
1
p1 + ap2
 Examples of ‘Kinky’ Solutions:
Perfect Complements
(a) p1x1* + p2x2* = m; (b) x2* = ax1*.
Substitution from (b) for x2* in
(a) gives p1x1* + p2ax1* = m
which gives * m * am
x1 = ; x2 =
p1 + ap2 p1 + ap2
 Examples of ‘Kinky’ Solutions:
Perfect Complements
(a) p1x1* + p2x2* = m; (b) x2* = ax1*.
Substitution from (b) for x2* in
(a) gives p1x1* + p2ax1* = m
which gives x * = m ; x * = am
1 2
p1 + ap2 p1 + ap2
 Examples of ‘Kinky’ Solutions:
Perfect Complements
x2 U(x1,x2) = min(ax1,x2)

*
x2 = x2 = ax1
am
p1 + ap 2
m
x*1 = x1
p1 + ap 2