0% found this document useful (0 votes)
36 views47 pages

Intermed 05

Download as pdf or txt
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 47

Intermediate Microeconomics — Week 5

Professor Boyd Sept. 20 & 22, 2022


QUIZ: Thursday’s quiz covers weeks 3 and 4, including sections 3.2
through section 4.3 (pp. 60–117).
4.4.11 A Non-Standard Budget Set: Quantity Discount 09/20/22
Suppose we have two goods, pizza and cell service, measured in gi-
gabytes (GB) per month. You have $60 income. Each pizza costs $10.
Cell service is $15 per GB for the first 2 GB, and $10 per GB for each
gigabyte past 2. The complication here is that the relative price changes,
which puts a kink in the budget frontier.
If you buy only pizza, you can afford 6 of them. Cell service is more
complicated. The first 2 GB cost a total of $30, leaving $30. At that
point, the price changes to $10 each. If you are buying only cell service,
you can afford $30 worth, which is 3 more GB, totalling 5 GB. If instead,
you spend the rest on pizzas, you can buy 3 pizzas and 2 GB. We plot
those three points, and then connect them.
Discounts and the Budget Set
b
6
px
/p y

5
=
3/

4
2
pizza

b
3
px
/p

2
y
=
1

b
0
0 1 2 3 4 5 6 7
cell service (GB)

The dashed lines show the budget frontier if the first 2 GB cost $10 per
GB, and additional GB cost $15 per GB.
2 INTERMEDIATE MICROECONOMICS

4.5 Consumer Choice


We’ve gathered the pieces of consumer theory, describing preferences
and their properties, and examining budget constraints.
Now we’re ready to put it all together and solve the consumer’s choice
problem: Maximizing utility subject to a budget constraint.
4. INTERMEDIATE MICROECONOMICS — WEEK 5 3

4.5.1 The Consumer’s Problem


The consumer knows their utility function and has income m. They face
prices px and py . How do they find the best point in the budget set?
How do they maximize utility?
We start by asking if any points that are strictly inside the budget set
(marked in red) can be best. We start with a point inside the budget
set, marked A in the graph. We add an indifference curve through A.
Completeness ensures there is such an indifference curve.

Is Bundle A Optimal?

6
5
4
3
2
1 A
b

0
0 1 2 3 4 5 6 7 8 9 10 11 12
4 INTERMEDIATE MICROECONOMICS

4.5.2 The Consumer’s Choice Must be on the Budget Frontier


The red hatched area is the budget set. By monotonicity, our consumer
prefers the consumption bundles in the area above and right of point A,
the blue hatched area, to A itself. The consumer can afford to improve
on A by choosing any point in the cross-hatched area.
Indeed, as the lower figure shows, any point above the indifference
through A and in the budget set (the cross-hatched part) is an improve-
ment.
No, Bundle A Is Not Optimal!

6
5
4
3
2
1 A
b

0
0 1 2 3 4 5 6 7 8 9 10 11 12

Whenever a consumption bundle is not on the budget frontier, we can


find such improvements. Conclusion: The best affordable consump-
tion bundle is on the budget frontier.
Bundle A is Really Not Optimal!

6
5
4
3
2
1 A
b

0
0 1 2 3 4 5 6 7 8 9 10 11 12
4. INTERMEDIATE MICROECONOMICS — WEEK 5 5

4.5.3 Finding the Best Bundle on the Frontier


Let’s examine a point on the budget frontier. Call it B. Now take the
indifference curve through B.
Can the consumer do better?
We start by marking the other point on the indifference curve through
B that is on the budget frontier. Call it C.

Is Bundle B Optimal?

6
5
4
3 b B

2
C
1 b

0
0 1 2 3 4 5 6 7 8 9 10 11 12
6 INTERMEDIATE MICROECONOMICS

4.5.4 Finding the Best Bundle on the Frontier


We color in the budget set (red) and the bundles that are better than B
(blue). This should make the situation clear.
The bundles on the budget frontier that lie between B and C are the
affordable bundles that are better than B. We start moving down and
right along the budget frontier to improve.

Bundle B is Still Not Optimal

6
5
4
3 b B

2
C
1 b

0
0 1 2 3 4 5 6 7 8 9 10 11 12

Another way of looking at this is that the slope of the indifference curve
through B is higher than the slope of the budget line. This means that
the marginal value of good X in terms of Y, the MRS, is higher than the
its price in terms of Y, px /py . So the consumer can improve by buying
more X and less Y.
4. INTERMEDIATE MICROECONOMICS — WEEK 5 7

4.5.5 The Consumer’s Choice


The best consumption bundle the consumer can afford is at point E.
From point E, there are no more points that we can afford (red) that
are also better than E (blue). That means that E is the best bundle that
our consumer can afford—the consumer’s optimum. This is the point in
the budget set with the highest possible utility. It maximizes utility.

Bundle E is Optimal!!

6
5
4
3
E
2 b

1
0
0 1 2 3 4 5 6 7 8 9 10 11 12
8 INTERMEDIATE MICROECONOMICS

4.5.6 Characterizing the Utility Maximum


The utility maximizing bundle E has two properties:
1. The bundle E is on the budget frontier
2. The budget frontier is tangent to the indifference curve at E. As a
result, the marginal rate of substitution at E is equal to the relative
price.
The point E = (x∗, y∗ ) = (5, 2) is our first point on the demand curve.
Its location depends on the consumer’s preferences (utility), income, and
the prices of both goods.

Optimality Conditions

6
5
4
3
y∗ 2 b E

1
0
0 1 2 3 4 5 6 7 8 9 10 11 12
x∗
4. INTERMEDIATE MICROECONOMICS — WEEK 5 9

4.5.7 The Two Optimality Equations


We restate the two conditions for optimality as two equations:

m = px x + py y Budget equation
px
MRS = Optimality equation.
py

Recall that
MUx px MUx
MRS = so = MRS = .
MUy py MUy
10 INTERMEDIATE MICROECONOMICS

4.5.8 An Alternative: Marginal Utility per Dollar


Let’s take a closer look at the optimality equation.

MUx px
= = MRS .
MUy py

Ignoring the MRS term, divide by px and multiply by MUy to obtain

MUy MUx [utils]/[x-unit] [utils]


= = = .
py px [$]/[x-unit] [$]

where the brackets indicate we are giving the units of each expression.
The MU /p terms are the marginal utility per dollar spent for each of
the goods. To maximize utility, we must spend each dollar in a way that
gives us the biggest bang per buck—the biggest marginal utility per
dollar.
Later, in section 4.5.24, we will see that we can easily include the
corner case.
4. INTERMEDIATE MICROECONOMICS — WEEK 5 11

4.5.9 Optimality by the Numbers I


The optimality graph uses the Cobb-Douglas utility function u(x, y) = xy.
Prices are px = 4 and py = 10. Income is $40. The relative price is then
px /py = 4/10 = 2/5 and the marginal rate of substitution is y/x. That
gives us two equations: the budget equation and the optimality equation
MRS = px /py . In other words,

4x + 10y = 40 Budget
y px
= = 2/5. Optimality
x py

Eliminating the common factor of 2 from the first equation and clearing
the denominators of the second, we find

2x + 5y = 20
(1)
5y = 2x.
12 INTERMEDIATE MICROECONOMICS

4.5.10 Optimality by the Numbers II

We solve system (1) to find (x∗, y∗ ). First, let’s repeat (1), emphasizing
the 5y terms.
2x + 5y = 20
(1)
5y = 2x.

Substitute 2x for 5y in the first equation of (1). That gives us

2x + 2x = 20 or 4x = 20.

Then x∗ = 5. We now plug this in the budget equation to find y∗ ,

2(5) + 5y = 20 or 5y = 10.

It follows that y∗ = 2. The optimal consumption point is (x∗ , y∗) =


(5, 2). This tells us both the demand for X and for Y at current prices
(px , py ) = (4, 10) and income m = 40.

The Optimal Consumption Point

6
5
4
3
(x∗ , y∗ )
2 b

1
0
0 1 2 3 4 5 6 7 8 9 10 11 12
4. INTERMEDIATE MICROECONOMICS — WEEK 5 13

4.5.11 Demand with Cobb-Douglas Utility I


To derive a demand function, we need to solve the consumer’s problem
for any price px . We consider the Cobb-Douglas Utility u(x, y) = xayb
with a, b > 0. In this case, the marginal utilities are

MUx = axa−1yb
MUy = bxa yb−1.

The marginal rate of substitution is

MUx axa−1yb ay
MRS = = a b−1 = .
MUy bx y bx
14 INTERMEDIATE MICROECONOMICS

4.5.12 Demand with Cobb-Douglas Utility II


Now

m = px x + py y Budget equation
ay px
= Optimality equation.
bx py

Multiplying the second equation by (py x),

m = px x + py y
a
py y = px x.
b

Then we can write


a
m= py y + py y
b
 a
= 1+ py y
 b 
a+b
= py y.
b
4. INTERMEDIATE MICROECONOMICS — WEEK 5 15

4.5.13 Demand and Cobb-Douglas Utility


Since  
a+b
m= py y,
b
we have
b
py y = m,
a+b
b
showing that consumer spends a share a+b of income on good y.
Demand for Y is then
 
∗ b m
y = .
a + b py

By the optimality equation


 
a m
x∗ = .
a+b px

The equations tell us not only demand for X and Y as a function of


their own prices, but also tell us how demand varies with the preference
parameters a and b, with prices of both goods px and py , and with
income. In this case, py does not affect demand for X.
16 INTERMEDIATE MICROECONOMICS

4.5.14 Graphing Cobb-Douglas Demand


To graph the demand for X, we must hold everything but px constant. For
this example, we set a = 1, b = 2, and m = 12 (py does not appear in
the demand equation, so it is irrelevant). Then demand D1 is x = 4/px ,
which is solely a function of px . Inverse demand is px = 4/x. If we
change income to m = 24, we get the demand curve D2 , x = 8/px .

px Two Cobb-Douglas Demand Functions

4
3 D2

2 D1

1
0 x
0 1 2 3 4 5 6 7 8 9 10 11 12

Changing preferences (a, b) or income (m) will shift demand. If we


change b to 4, holding m = 12, we get x = 2.4/px , a decrease in the
demand for X (not illustrated).
Notice that py does not affect the demand for X.
4. INTERMEDIATE MICROECONOMICS — WEEK 5 17

4.5.15 Demand with Perfect Complements I


We start with the utility function u(x, y) = min{ax, by} where a, b > 0.
This means that goods X and Y are perfect complements in a ratio of a/b
units of Y for every unit of X.
As we saw in section 4.3.16, there is an issue about how to interpret the
marginal rate of substitution for such utility functions. When ax > by,
we are on the horizontal part of some indifference curve and MRS = 0.
When ax < by, we are on the vertical portion of some indifference
curve, and MRS = +∞. It is only when ax = by that the tangent lines
can have finite, non-zero slopes. As the figure below illustrates, tangents
with any slope are possible there.

Perfect Complements
y
MRS = +∞

4
3

2
MRS = 0
1 b

0 x
0 1 2 3 4 5 6 7 8
18 INTERMEDIATE MICROECONOMICS

4.5.16 Demand with Perfect Complements II


This slope problem means that the only place that the optimality con-
dition can be satisfied is along the line ax = by. We again have two
equations:

m = px x + py y Budget equation
ax = by Optimality equation.

Then y = (a/b)x, and the budget equation becomes


 
a bpx + apy
m = px x + py x = x
b b

so
bm
x∗ = .
bpx + apy
Using y = (a/b)x yields

am
y∗ = .
bpx + apy
4. INTERMEDIATE MICROECONOMICS — WEEK 5 19

4.5.17 Graphing Demand with Perfect Complements


To graph the demand for X, we must hold everything but px constant.
We do that by setting a = b = 1, py = 2, and m = 15. Then demand
is x = 15/(2 + px ). Inverse demand is px = 15/x − 2.

A Leontief Demand Function


px

7
6

4
3

1 D
0
0 1 2 3 4 5 6 7 8 9 x
bm
apy

Changing preferences (a, b), income (m), or the price of the comple-
mentary good (py ) will shift the demand curve.
20 INTERMEDIATE MICROECONOMICS

4.5.18 Demand with Perfect Substitutes I


Perfect substitutes can be described by linear utility functions of the form
u(x, y) = ax + by, where a, b > 0. The marginal rate of substitution is
MRS = a/b. The optimality condition requires that

a px
= MRS = .
b py

Since a and b are constants that describe preferences, there is only one
relative price where the optimality condition holds. At any other relative
prices, the optimality condition can never be satisfied. We’ll look more
at that in a bit. When optimality holds, the budget line coincides with
the highest indifference curve that touches the budget set. Every point
on the budget frontier is equally good, and maximizes utility.

y Perfect Substitutes: MRS Equal to Relative Price I

6
5
4
3
2
1
0 x
0 1 2 3 4 5 6 7 8 9 10 11 12

On the diagram, the budget line is the thick red line and the maximizing
indifference curve is shown as black, superimposed on the budget line.
Any point on the budget line is an optimal point.
4. INTERMEDIATE MICROECONOMICS — WEEK 5 21

4.5.19 Demand with Perfect Substitutes II


We repeat the previous diagram, with the entire budget set hatched red
and the consumption bundles with higher utility than the budget frontier
hatched blue, allowing you to see the division into better bundles (blue),
which the consumer cannot afford, and worse bundles (red hatched),
which the consumer can afford. These regions are separated by the
budget frontier/indifference curve.

y Perfect Substitutes: MRS Equal to Relative Price II

6
5
4
3
2
1
0 x
0 1 2 3 4 5 6 7 8 9 10 11 12
22 INTERMEDIATE MICROECONOMICS

4.5.20 Demand with Perfect Substitutes III


What happens if the budget frontier and indifference curves don’t align?
This happens in the diagram below, where the budget line is again shown
in red.

y Perfect Substitutes: MRS Greater than Relative Price I

6
5
4
3
2
A
1 b

0 x
0 1 2 3 4 5 6 7 8 9 10 11 12

Although the point A is on the budget frontier, it fails optimality. We


know the optimal point is on the budget frontier. Where is it? Which
way should we move along the budget frontier to increase utility.
4. INTERMEDIATE MICROECONOMICS — WEEK 5 23

4.5.21 Demand with Perfect Substitutes IV


Here the relative price of X, px /py , is greater than the marginal rate of
substitution. The relative price is the absolute slope of the red line, the
budget frontier. This is clearly higher than marginal rate of substitution,
the absolute slope of the black line, the budget frontier.
This means that the marginal value of X (MRS) is less than its cost (rela-
tive price), so the consumer benefits from cutting back on consumption
of X
Our consumer needs to reduce purchases of X and increase purchases
of Y. This will move them up to the left along the budget frontier.

y Perfect Substitutes: MRS Greater than Relative Price II

6
5
4
3
2
A
1 b

0 x
0 1 2 3 4 5 6 7 8 9 10 11 12

How far left do we have to move?


24 INTERMEDIATE MICROECONOMICS

4.5.22 Left Corner Solution


We move left as far as we can, until we can’t reduce our consumption of
X any more. That doesn’t happen until we hit the Y-axis, where x = 0.

y Left Corner Solution

6
5
4 b (x∗ , y∗)

3
2
1
0 x
0 1 2 3 4 5 6 7 8 9 10 11 12

As you can see on the diagram, the highest indifference curve that
touches the budget line is the one through (x∗, y∗ ) = (0, 4), making it the
optimal choice.
When the relative price is always greater than the MRS, the optimal
consumption bundle will be the vertical intercept of the budget frontier.
This solution at the corner of the budget set is called a corner solution.
4. INTERMEDIATE MICROECONOMICS — WEEK 5 25

4.5.23 Right Corner Solution


When the relative price is less than the (constant) marginal rate of substi-
tution, the situation reverses. We move right as far as we can, not left.
We stop when we can’t reduce our consumption of Y any more. That
doesn’t happen until we hit the X-axis, where y = 0.

y Right Corner Solution

6
5
4
3
2
1 (x∗, y∗ )
0 b
x
0 1 2 3 4 5 6 7 8 9 10 11 12

As you can see on the diagram, the highest indifference curve that
touches the budget line is the one through (x∗ , y∗ ) = (10, 0), making it
the optimal choice.
When the relative price is always less than the MRS, the optimal con-
sumption bundle will be the horizontal intercept of the budget frontier.
26 INTERMEDIATE MICROECONOMICS

4.5.24 Corner Solution Summary


The right corner of the budget set, (m/px , 0) maximizes utility whenever

px
MRS(m/px , 0) ≥
py

at the corner. When MRS diminishes, it must be higher everywhere


to the left of the corner, meaning that everywhere else on the budget
frontier
px
MRS > .
py
This means that the usual optimality condition that MRS = px /py cannot
be satisfied.
Similarly, if the left corner obeys

px
MRS(0, m/py ) ≤ ,
py

a diminishing marginal rate of substitution means that everywhere else


on the budget frontier
px
MRS < .
py
Again, the usual optimality condition fails away from the corner. In this
case, the left corner maximizes utility.
4. INTERMEDIATE MICROECONOMICS — WEEK 5 27

4.5.25 Consumer Choice with a Plethora of Goods


The bang for the buck formulation has the advantage that it applies
no matter how many goods the consumer considers for purchase. The
goods bought all have the same marginal utility per dollar. Any good
where the marginal utility per dollar at zero consumption is lower than
that common value will not be bought.
This is extremely helpful when dealing with fact that in modern
economies, most consumers do not purchase any the vast majority of
goods that are available. If you do not believe this, take a trip to your
favorite supermarket, and starting counting the number of goods avail-
able to estimate the total (count each brand and variety separately, don’t
just say “bread”). Then ask yourself what fraction of those goods you, or
even your family, have ever bought.
28 INTERMEDIATE MICROECONOMICS

4.5.26 Graphing Demand with Perfect Substitutes I


To graph the demand for X, we again hold everything but px constant.
To do this, we set a = 3, b = 1 (MRS = 3), py = 1, and m = 9. The
key price px , where we switch from one corner to the other, is when
MRS = px /py , meaning px = (a/b)py . For us, (a/b)py = 3.
When px > (a/b)py = 2, we consume no X. When px < (a/b)py = 2,
we spend everything on X, so x = m/px = 6/px .
When px = (a/b)py = 2, we can consume any amount of X we can
afford, anything from 0 to m/2 = 3.

Demand under Perfect Substitutes


px

7
6

4
(a/b)py 3

2
D
1
0
0 1 2 3 4 5 6 7 8 9 x
4. INTERMEDIATE MICROECONOMICS — WEEK 5 29

4.5.27 Graphing Demand with Perfect Substitutes II


Here we increase the price of Y from py = 1 to p′y = 5/3. Notice the
change in demand.

Demand Increase under Perfect Substitutes


px

7
6
(a/b)p′y 5

4
D′
(a/b)py 3

2
D
1
0
0 1 2 3 4 5 6 7 8 9 x

See how demand switches from the one curve to the other when the
price of X reaches the critical level (a/b)py .
Demand again depends on prices, the preference parameters, and
income.
30 INTERMEDIATE MICROECONOMICS

4.5.28 Summary of Calculated Demand Functions


Cobb-Douglas utility, u(x, y) = xa yb :
 
∗ a m
x = .
a+b px

Perfect Complements, u(x, y) = min{ax, by}:

bm
x∗ = .
bpx + apy

Perfect Substitutes (linear), u(x, y) = ax + by:


0 when px > (a/b)py

x∗ = [0, pmx ] when px = (a/b)py
 m
px when px < (a/b)py
4. INTERMEDIATE MICROECONOMICS — WEEK 5 31

4.5.29 Price Elasticities for Cobb-Douglas Demand


We can take a quick look at elasticities for Cobb-Douglas demand
 
∗ a m
x = .
a+b px

Spending on X is
∗ a 

px x = m.
a+b
Since this is independent of px , demand is unit elastic. The price elasticity
of demand is −1.
Demand is independent of py , so the cross-price elasticity of demand
is 0.
32 INTERMEDIATE MICROECONOMICS

4.5.30 Income Elasticity for Cobb-Douglas Demand


Finally, we consider income elasticity. To simplify the calculation, let
z = a/px (a + b), so that x∗ = zm. Since z doesn’t depend on income,
dx∗ /dm = z. Then

m dx∗ m
EI = ∗ = × z = +1.
x dm zm

This indicates that X is a normal good, and is right at the borderline of


being a superior good (recall that superior means EI > 1). The share of
income spent on X is
px x∗ a
= .
m a+b
4. INTERMEDIATE MICROECONOMICS — WEEK 5 33

4.5.31 Price Elasticity for Perfect Complements 09/22/22


Demand is
bm
x∗ = .
bpx + apy

To find the price elasticity of demand, we first take the derivative with
respect to px :

∂x∗ bm
= −b
∂px (bpx + apy )2
  
−b bm
=
bpx + apy bpx + apy
−b
= x∗ .
bpx + apy

Then
px ∂x∗ bpx
|Ed | = = <1
x∗ ∂px bpx + apy
Demand is always inelastic.
34 INTERMEDIATE MICROECONOMICS

4.5.32 Cross-Price and Income Elasticity for Perfect Complements


Demand is
bm
x∗ = .
bpx + apy

To find the cross-price elasticity of demand, we first take the derivative


with respect to py :

∂x∗ bm
= −a
∂py (bpx + apy )2
  
−a bm
=
bpx + apy bpx + apy
−a
= x∗ .
bpx + apy

Then
py ∂x∗ −apy
Exy = = < 0.
x∗ ∂py bpx + apy
This is always negative, indicating that X and Y are complements.
Since x∗ is again proportional to m, its income elasticity is again +1.
5. INTERMEDIATE MICROECONOMICS — WEEK 5 35

5.1 Chapter 5: Individual and Market Demand


In this chapter we take a closer look at demand, focusing on general
characteristics rather than specific demand curves.
We know that demand describes the relation between the price of
a product and the quantity demanded—the quantity buyers wish to
purchase—all while holding other relevant variables constant. At the
end of the last chapter, we examined the effects of some of these other
variables—income, prices of other goods, preference parameters.
Quantity demanded depends on many variables, not just price. One
of these variables, income, is tied up with demand itself. This is because
changes in the price of X lead to changes in real income.
36 INTERMEDIATE MICROECONOMICS

5.1.1 Price and Real Income


We earlier considered the budget set with income $10, py = $5 and
three different prices for X: px = $1, p′x = $2, and p′′x = $3.33.
As the price of X increases, the budget line rotates clockwise about the
Y anchor point at (0, 2).

y Price Increase and Budget Set

3
m/py 2

1
0 x
0 1 2 3 4 5 6 7 8 9 10
m/p′′x m/p′x m/px

The budget set shrinks as the price of X increases, and expands as the
price decreases. Since there are more options available at lower prices,
and fewer at higher prices, we can say that real income (the consumer’s
consumption possibilities) increased as prices fell, and shrank as prices
rose.
When we change the price of X, it doesn’t make sense to think about
income being unchanged. Even when the dollar income is unchanged,
real income still changes. The effect due to changing income is called
the income effect. We will need to take the income effect into account
when analyzing demand.
5. INTERMEDIATE MICROECONOMICS — WEEK 5 37

5.1.2 The Engel Curve


We use an Engel curve to focus on the effect of income. The Engel curve
graphs quantity demanded as a function of income rather than price.1
All other variables, including price, are held constant.
For the demand functions we studied in detail, quantity demanded is
proportional to income, as illustrated below.

An Engel Curve
5
quantity demanded

4
3

1
0
0 1 2 3 4 5 6 7 8 9
income
A positive slope corresponds to a normal good (EI > 0), while a nega-
tive slopes indicates the good is inferior (EI < 0). Straight lines through
the origin indicate the income elasticity EI = 1.

1
Ernst Engel (1821–1896) was a German statistician who was the first to systematically study the
relationship between expenditures and income. He discovered that consumer spend a smaller share of
their on food as income increases. This is known as Engel’s Law.
38 INTERMEDIATE MICROECONOMICS

5.1.3 Unit Income Elastic Engel Curves


The diagram illustrates some unit income elastic Engel curves. Engel
curves for luxury goods will cross the unit elastic lines that lie to the left of
them, while goods with income elasticity less than one cross unit elastic
lines lying to the right.

A Gaggle of Engel Curves


quantity demanded

1
0
0 1 2 3 4 5 6 7 8
income
5. INTERMEDIATE MICROECONOMICS — WEEK 5 39

5.1.4 Engel Curve of a Superior Good


For example, the red Engel curve below keeps crossing unit elastic Engel
curves to the left of it. It must be a superior good.

Engel Curve of a Superior Good


quantity demanded

4
3

1
0
0 1 2 3 4 5 6 7 8
income
40 INTERMEDIATE MICROECONOMICS

5.1.5 Income Expansion Paths


A second way of portraying income effects is the income expansion
path (IEP). The income expansion path is graphed in X−Y space. It is
the locus of optimal consumption choices as income changes, for fixed
prices and a fixed indifference map.
Below, we use a Cobb-Douglas indifference map and prices px = py =
1. The dots indicate the optimal choices for incomes of $2, $4, $6, $8,
$10, and $12.

An Income Expansion Path


y
IEP

6 b

5 b

4 b

3 b

2 b

1 b

x
0
0 1 2 3 4 5 6 7 8
5. INTERMEDIATE MICROECONOMICS — WEEK 5 41

5.1.6 Interpreting Income Expansion Paths


The income expansion path is positively sloped at some income level
whenever both goods are normal there. If the slope is negative, one of
the goods must be inferior. To tell which, you have to follow the path as
income increases. If the path moves toward the vertical axis, X is inferior.
If the path moves toward the horizontal axis, Y is inferior.
The diagram shows a portion of an income expansion path where X is
inferior. If the arrow were reversed, Y would be inferior.

y Income Expansion Path: X is Inferior

5
4
3
2
1
0 x
0 1 2 3 4 5 6 7 8 9
42 INTERMEDIATE MICROECONOMICS

5.1.7 Expenditure Minimization I


To deal with the issue of changes in real income, we turn the utility
maximization problem on its head. Instead of maximizing utility over a
budget set, we try to minimize cost over the set of better consumption
bundles. To do this, we start with a utility level u and consider all
consumption bundles that give the consumer at least utility u.

y Bundles with Utility at Least u

6
5
4
3
2
1 u
0 x
0 1 2 3 4 5 6 7 8 9 10 11 12

Then we try to find the cheapest such consumption bundle, given


prices px and py , rather than the most valuable consumption bundle, as
with utility maximization. Here we try to minimize consumer expendi-
ture, the cost to the consumer.
This sort of trick, where we replace maximization problems by closely
related minimization problems, and vice-versa, is one of the most useful
in the microeconomist’s toolbox. There are many variations on it and
we could spend the rest of the semester exploring these variations. We
don’t have time for that, so we will just focus on one such problem, cost
minimization.
5. INTERMEDIATE MICROECONOMICS — WEEK 5 43

5.1.8 Expenditure Minimization II


Let px = 2 and py = 6, yielding a relative price of 1/3. If we take any line
of slope −1/3, all of the consumption bundles on that line cost the same.
We refer to these budget-like lines as isocost lines or isoexpenditure
lines. The expenditure associated with an isoexpenditure line is px times
its horizontal intercept, or py times the vertical intercept.
It’s pretty clear that we can do better than the top two isoexpenditure
lines by moving to a lower isoexpenditure line. The bottom two do not
give us enough utility, and the middle one comes very close to solving
the problem. It almost does.

y Isoexpenditure Lines and Bundles with Utility at Least u

6
5
4
3
2
1 u
0 x
0 1 2 3 4 5 6 7 8 9 10 11 12
44 INTERMEDIATE MICROECONOMICS

5.1.9 Cost Minimization: Solution


As you’ve problem guessed from the previous diagram, the solution in-
volves a tangency between the isoexpenditure line and the indifference
curve. It also has to yield utility u.

MUx px
MRS(x, y) = = Optimality equation.
MUy py
u = u(x, y) Utility constraint
√ √
The expenditure minimizing point is (x∗ , y∗) = ( 30, 30/3 , which

√ is
approximately
√ √(5.48, 1.83) and the corresponding expenditure is 2 30+
6 30/3 = 4 30 ≈ 21.91.

y Minimizing Expenditure to Obtain Utility u

6
5
4
3
2 b

1 u
0 x
0 1 2 3 4 5 6 7 8 9 10 11 12
5. INTERMEDIATE MICROECONOMICS — WEEK 5 45

5.1.10 Minimum Expenditure Function


The minimum expenditure we calculated depends on prices px and
py , and on the utility level u. We use the notation e(px , py , u) for the
minimum expenditure.
Here e stands for expenditure. We prefer to not use the word “cost”
here since we want to reserve the term “cost function” for firms, not
consumers. Using different terms helps avoid confusion. In fact, a firm’s
cost function will be calculated the same way. We will see how to do the
same kind of calculation later for firms.
46 INTERMEDIATE MICROECONOMICS

5.1.11 The True Cost-of-Living Index


We can use the expenditure function to define a price index, the true
cost-of-living index. Here we have two sets of prices, (px , py ) and
(p′x , p′y ). We want to measure how much prices have risen or fallen.
If both prices increase by the same percentage, its easy. Just use that
percentage. In the real world, with many prices, this just doesn’t happen.
When prices in general rise, some prices may actually go down, some go
up, some go up a lot. In August 2022, the price of food went up 0.8%
while the price of gasoline went down 12.2%. How can we determine
what happened to prices in general? We use a price index to measure
this.2
We’re interested in how price changes affect consumers. The most
natural measure is to ask how much a consumer’s income would have to
rise to be able to maintain their standard of living, their utility level. We
know it takes income of e(px, py , u) to get utility u at prices (px , py ) and
that it takes income e(p′x, p′y , u) to get the same utility at prices (p′x , p′y ).
So we just take the ratio. The true cost-of-living index P is defined by3

e(p′x , p′y , u)
P= .
e(px , py , u)

2
One commonly used price index is the consumer price index (CPI). The CPI was virtually unchanged
in August. However, it’s not a good measure of month-to-month changes because two parts of it, food
and energy prices, are both highly volatile. A better measure for month-to-month changes is the price
of all items except food and energy. The seasonally adjusted version should be slightly better still. That
went up 0.6% in August, for an annual rate of about 7.4%. Inflation works like compound interest, so
the calculation is (1 + 0.6%)12 = (1.006)12 ≈ 1.0744 = 1 + 7.44%.
3
The true cost-of-living index was introduced by Alexander A. Konüs in 1924.
5. INTERMEDIATE MICROECONOMICS — WEEK 5 47

5.1.12 Price Indices in Practice 09/27/22


The problem with the true cost-of-living index is twofold. We can’t
observe utility levels, and different consumers have different preferences
(implying different price indices).
We can observe consumption bundles, or at least average consump-
tion bundles. The consumer price index uses an average consumption
bundle. We measure its cost at the old prices and new, and then take the
ratio to get the consumer price index. This is roughly how the consumer
price index is calculated, but sometimes involves refinements that get
us closer to the true cost-of-living index. Some of these, such as the
treatment of housing prices are controversial among the public, but are
justified by economic theory. (This is not to say they are perfect.)
The consumption bundle is updated frequently so that it will approxi-
mate the cost minimizing bundle.
The Bureau of Labor Statistics calculates various price indices. As
part of that calculation they have to figure out an average consumption
bundle. They collect data on all sorts of consumer expenditures. A
recent example can be found at
https://www.bls.gov/news.release/pdf/cpi.pdf
Check out the tables near the end of the pdf to see how spending is
broken down.
September 22, 2022

Copyright c 2022 by John H. Boyd III: Department of Economics, Florida International Univer-
sity, Miami, FL 33199

You might also like