Course Manual 10.1 and 10.2
Jacques 5.5 and 5.6
The behaviour of economic actors is often constrained by the economic resources they have at their disposal
Example:
–Individuals maximising utility will be subject to a budget constraint
–Firms maximising output will be subject to a cost constraint
The function we want to maximise is called the objective function
The restriction is called the constraint
Max U = f(X1 X2) = X21 X2
[ X1* X2* ]
Subject to
g(X1 X2)= P1X1 + P2X2 - M = 0
Three Ways to do this:
- By Substitution
- Economic Theory
- Lagrange Multiplier
Method 1: By Substitution
Max the Objective function:
Max U = f (X1 X2) = X21 X2
[ X1* X2* ]
Subject to the constraint:
P1X1 + P2X2= M
Step 1:Use the constraint to express X2 in terms of X1 (or vice-versa)
Step 2:
Substitute into objective function
Step 3:
F.O. Condition
( P1X1 = 2/3M, expenditure on good 1 is 2/3 of income)
S. O. Condition
For a Max,
X1 needs to be large enough to sign N.D.
How Large? First find the X1 that sets,
Answer:
The optimal f11< 0
Step 4: Substitute into constraint to find corresponding value of X*2 that maximises objective function
Since P1X1 + P2X2= M
(note, rearranging, P2X2 = 1/3 M . expenditure on good 2 is 1/3 of M)
Method 2: By Economic Theory
Max U = f (X1 X2) =
[ X1* X2* ]
S.t. g (X1 X2)= P1X1 + P2X2 - M =0
- Total differential of the objective function f (X1 X2):
dU = f1 dX1+ f2 dX2 = 0
(f1f2, marginal utility of X1X2, respectively)
eq.1
- Total differential of the constraint:
g(X1X2) = P1 X1+ P2 X2 – M= 0
P1 dX1+ P2 dX2 = 0
eq.2
Condition 1 – from eq1 and eq2
Condition 2
From condition 1, substitute P1X1 (or P2X2) into the constraint, and solve
Hence,
Method 3: By The Lagrange Multiplier
Max U = f (X1 X2) =
[ X1* X2* ]
s.t.
g (X1 X2)= P1X1 + P2X2- M = 0
Step 1: Define the Lagrangean Function L (objective function + constraint)
Max L = f(X1 X2 ) + g(X1 X2)
[X1* X2* *]
Max L = X12X2 + (M – P1X1– P2X2 )
[X1* X2* *]
OR L = X21X2 – (P1X1 + P2X2 –M)
Step 2: First Order Conditions
Set dL= L1.dX1 + L2.dX2 = 0
1. L1 = 2X1X2 – P1 = 0 eq1
2. L2 = X12 – P2 = 0 eq2
Set g(X1X2) = 0
3. L = M – P1X1 – P2X2 = 0 eq3
Step 3: Solve the system of equations
Solving equations 1 & 2:
2X1P2X2 = P1X12
2P2X2 = P1X1
And substituting into eq 3
P1X1 + P2X2 – M= 0
2P2X2+ P2X2 – M= 0
from eq 3:
Substituting in for X2:
(again, note that rearranging reveals that P1X! = 2/3 M and P2X2 = 1/3 M .2/3 of income spent on good 1, and 1/3 on good2)
Step 4: Second Order Condition
d2L = L11.dX21+L12.dX1dX2+L21.dX2dX1+L22dX22
s.t. g1.dX1 + g2.dX2= 0
or dX2= -(g1/ g2).dX1
N. D. for a Max
d2L=[L11.g22 - 2L12.g1.g2+L22g21]dX21/g22
d2L= ΦdX21/g22 , if Φ < 0, N. D.
= 0+ P1(P2.2X1) -P2(-P1.2X1+ P2.2X2)
= 4P1P2X*1 - 2P2P2.X*2
= 4P2(2/3. M) - 2P2(1/3.M)
= (8P2M - 2P2M)/3
= 2P2M > 0
d2L < 0, N. D. (Max)
A consumers preferences can be represented by the Utility Function, U(x,y)=x.y. How much will the utility maximising consumer demand of goods x and y if they have an income of €100, the price of good x is €5 and the price of good y is €1?
Lagrangean Method:
L = x.y + [100 – 5x – y]
Eq. 1. LX = y – 5= 0
Eq. 2. Ly = x – = 0
Eq. 3. L =100 – 5x – y = 0
Solving
Eq 1&2: = y/5 = x
So y* = 5x
Substitute into eq.3:
100 = 5x + y = 10x
So x* = 10
Then y*= 5x=50
And U* = xy = 500
(note that P1X1 = ½ M and P2X2 = ½ M .
½ of income spent on good 1, and ½ on good2)
Second Order Condition:
Topic 10: Q1
Maximise (i) U = x½y½ (ii) U = x.y (iii) U = logx + logy. Let price x = p and price of y = q, and m = income
(i) L = x½y½ + [m – px – qy]
Eq. 1. Lx = ½ x-½y½ - p= 0
Eq. 2. Ly = ½ x½y-½ - q = 0
Eq. 3. L = m – px – qy = 0
Eq 1&2: =
px½y-½ = qx-½y½
p x½/ x-½ = q y½/ y-½
px = qy….expenditure on good x = expenditure on good y
Substitute into eq.3: m = px + qy = 2px
px* = ½ m qy* = ½ m
Cobb-Douglas: if U = xayb and a+b=1 (a= ½ & b= ½ above)Then income expenditure share on x is a and on y is b.
(ii) L = xy + [m – px – qy]
Same Solution: x* = m/2p (and px = ½ m) and y*= m/2q (and qy = ½ m)
Cobb-Douglas: if U = xayb and a+b1 (a=1 & b=1 above)
Then income expenditure share on x is a/a+b and on y is b/a+b
(iii) L = log x + log y + [m – px – qy]
Same Solution: x* = m/2p (and px = ½ m) and y*= m/2q (and qy = ½ m)
Note: log x + log y = x.y
3 utility functions - same ranking of preferences – represent monotonic transformations of a utility function.