Chari Boru's - DR - Haasan Huseen - Microeconomics I PPT (PHD) Program

Bulehora University
Microeconomics I (PECOC711)
Lecture Slide
By
Hassen Beshir (PhD)
Associate professor in Applied and Agricultural

Economics
January 2022
1
Introduction:
Basic Principles
2
Theoretical Models
 Economists use models to describe
economic activities
 While most economic models are

abstractions from reality, they provide aid
in understanding economic behavior
3
Verification of Economic Models
 There are two general methods used to
verify economic models:
Direct approach
 Establishes the validity of the model’s assumptions
Indirect approach
 Shows that the model correctly predicts real-world
events
4
Verification of Economic Models
 We can use the profit-maximization model to
examine these approaches
◦ is the basic assumption valid? do firms really seek to
maximize profits?
◦ can the model predict the behavior of real-world
firms?
5
Features of Economic Models
 Ceteris Paribus assumption
 Optimization assumption
 Distinction between positive and

normative analysis
6
Ceteris Paribus Assumption
 Ceteris Paribus means “other things the same”

 Economic models attempt to explain simple
relationships
◦ focus on the effects of only a few forces at a time
◦ other variables are assumed to be unchanged
during the period of study
7
Optimization Assumptions
 Many economic models begin with the
assumption that economic actors are
rationally pursuing some goal
◦ consumers seek to maximize their utility
◦ firms seek to maximize profits (or minimize
costs)
◦ government regulators seek to maximize
public welfare
8
Optimization Assumptions
 Optimization assumptions generate
precise, solvable models
 Optimization models appear to be

perform fairly well in explaining reality
9
Positive-Normative Distinction
 Positive economic theories seek to
explain the economic phenomena that is
observed
 Normative economic theories focus on

what “should” be done
10
The Economic Theory of Value
 Early Economic Thought
◦ “value” was considered to be synonymous
with “importance”
◦ since prices were determined by humans, it
was possible for the price of an item to differ
from its value
◦ prices > value were judged to be “unjust”
11
 The Founding of Modern Economics
◦ the publication of Adam Smith’s The Wealth of Nations
is considered the beginning of modern economics
◦ distinguishing between “value” and “price” continued
(illustrated by the diamond-water paradox)
 the value of an item meant its “value in use”
 the price of an item meant its “value in exchange”
12
 Labor Theory of Exchange Value
◦ the exchange values of goods are determined by what
it costs to produce them
 these costs of production were primarily affected by labor
costs
 therefore, the exchange values of goods were determined by
the quantities of labor used to produce them
◦ producing diamonds requires more labor than
producing water
13
 The Marginalist Revolution
◦ the exchange value of an item is not determined by
the total usefulness of the item, but rather the
usefulness of the last unit consumed
 because water is plentiful, consuming an additional unit has a
relatively low value to individuals
14
 Marshallian Supply-Demand Synthesis
◦ Alfred Marshall showed that supply and demand
simultaneously operate to determine price
◦ prices reflect both the marginal evaluation that
consumers place on goods and the marginal costs of
producing the goods
 water has a low marginal value and a low marginal cost of
production  Low price
 diamonds have a high marginal value and a high marginal
cost of production  High price
15
Supply-Demand Equilibrium
Price
Equilibrium S
QD = Q s The supply curve has a positive
slope because marginal cost
rises as quantity increases
P*
The demand curve has a

negative slope because
the marginal value falls as
D
quantity increases
Q* Quantity per period
16
qD = 1000 - 100p
qS = -125 + 125p
Equilibrium  qD = qS
1000 - 100p = -125 + 125p
225p = 1125
p* = 5
q* = 500
17
 A more general model is

qD = a + bp
qS = c + dp
Equilibrium  qD = qS
a + bp = c + dp
ac
p* 
d b
18
A shift in demand will lead to a new equilibrium:

Q’D = 1450 - 100P
Q’D = 1450 - 100P = QS = -125 + 125P
225P = 1575
P* = 7
Q* = 750
19
Price An increase in demand...
S
…leads to a rise in the
equilibrium price and
7 quantity.
5
D’
D
500 750 Quantity per period
20
 General Equilibrium Models
◦ the Marshallian model is a partial equilibrium
model
 focuses only on one market at a time
◦ to answer more general questions, we need a
model of the entire economy
 need to include the interrelationships between
markets and economic agents
21
 The production possibilities frontier can
be used as a basic building block for
general equilibrium models
 A production possibilities frontier shows
the combinations of two outputs that can
be produced with an economy’s
resources
22
A Production Possibility Frontier
 The production possibility frontier
reminds us that resources are scarce
 Scarcity means that we must make
choices
◦ each choice has opportunity costs
◦ the opportunity costs depend on how much
of each good is produced
23
 Suppose that the production possibility
frontier can be represented by
2 x 2  y 2  225
• To find the slope, we can solve for Y
y  225  2 x 2
• If we differentiate
dy 1 2 1 / 2  4 x  2x
 (225  2 x )  ( 4 x )  
dx 2 2y y
24
dy 1 2 1 / 2  4 x  2x
 (225  2 x )  ( 4 x )  
dx 2 2y y
 when x=5, y=13.2, the slope= -2(5)/13.2= -0.76

 when x=10, y=5, the slope= -2(10)/5= -4
 the slope rises as y rises
25
 Welfare Economics
◦ tools used in general equilibrium analysis have been
used for normative analysis concerning the
desirability of various economic outcomes
 economists Francis Edgeworth and Vilfredo Pareto helped
to provide a precise definition of economic efficiency and
demonstrated the conditions under which markets can
attain that goal
26
Modern Tools
 Clarification of the basic behavioral

assumptions about individual and firm
behavior
 Creation of new tools to study markets
 Incorporation of uncertainty and imperfect
information into economic models
 Increasing use of computers to analyze data
27
Important Points to Note:
 Economics is the study of how scarce

resources are allocated among alternative
uses
◦ economists use simple models to understand
the process
28
 The most commonly used economic

model is the supply-demand model
◦ shows how prices serve to balance
production costs and the willingness of
buyers to pay for these costs
29
 The supply-demand model is only a

partial-equilibrium model
◦ a general equilibrium model is needed to look
at many markets together
30
 Testing the validity of a model is a

difficult task
◦ are the model’s assumptions reasonable?
◦ does the model explain real-world events?
31
Chapter two-1:
PREFERENCES AND UTILITY
32
PREFERENCES AND UTILITY (Cont…)
 The basic economic model o of consumer

behavior is that people choose the best things they
can afford. But, what is ‘best’?
 Consumption bundles: The objects of consumer
choice
 Consumption set: The set of consumption bundles,
(X): a complete list of goods and services
involved in the choice problem. It is important to
include all appropriate goods in the definition of
the consumption bundle.
33
 For simplicity, two goods assumption (one

good and all other goods) often results in
useful insights, i.e. (x1, x2)
 Weak preference: x1 ≽ x2
 Strict preference: x1 ≻ x2
 Indifference: x1 ~ x2
34
 If the consumer prefers one bundle to

another, it means that he would choose
one over the other. Thus, the idea of
preference is based on the consumer’s
choice. In order to tell whether one bundle
is preferred to another, we see how the
consumer behaves in choice situations
involving the two bundles.
35
Fundamental assumptions about preferences (Axioms of
rational choice)
 Completeness: For all x and y, in X, either x ≽ y, or x ≼ y,

or both. This assumption means that any two bundles can be
compared and that individuals are assumed not to be
paralyzed by indecision. They completely understand and
can always decide about the desirability of two choices. It
also rules out the possibility that x is preferred to y and y is
preferred to x.
◦ if A and B are any two situations, an individual can always specify

exactly one of these possibilities:
 A is preferred to B
 B is preferred to A
 A and B are equally attractive
36
rational choice)
 Transitivity: For all x , y, and z in X, if x ≽ y and y ≽ z,
then x ≽ z. This assumption states that the individual’s
choices are internally consistent. This assumption may be
violated when the consumer does not fully understand the
consequences of the choices he is making. If we assume
that choices are fully informed, the transitivity property is a
reasonable assumption. This assumption is necessary for
any discussion of preference maximization, i.e. If we have
to have a theory where people are making ‘best’ choices,
preferences must satisfy the transitivity axiom.
◦ if A is preferred to B, and B is preferred to C, then A is preferred to
C
◦ assumes that the individual’s choices are internally consistent
37
rational choice)
 Continuity: For all y in X, the sets {x: x ≽ y} and {x: x ≼
y} are closed sets. In other words, for all bundles x, the
better set and the worse sets are closed. It then follows that
{x: x ≺ y} and (x: x ≻ y} are open sets. The assumption is
necessary to rule out certain discontinuous behavior. The
consequence of this assumption is that if y is strictly
preferred to z, and if x is a bundle close enough to y, then x
must be strictly preferred to z. This assumption is required
if we wish to analyze individual’s responses to relatively
small changes in income and prices.
◦ if A is preferred to B, then situations suitably “close to” A must also
be preferred to B
◦ used to analyze individuals’ responses to relatively small changes in
income and prices
38
Fundamental assumptions about preferences
(Axioms of rational choice)
 Weak Monotonicity: if x ≥ y, then x ≽ y.
 Strong monotonicity: if x >y, then x ≻ y.
 Local non-satiation: Given any x in X and any ε > 0,
then there is some bundle y in X with |x-y| < ε such that
y ≻ x.
◦ Economic goods: Whatever economic quantities they represent,
more of any particular good during some period is preferred to
less.
◦ Local non-satiation says that one can always do a little better
even if one is restricted to only small changes in the consumption
bundle. Local non-satiation rules out thick indifference curves.
39
Other assumptions about Preference
Two assumptions that are often used to guarantee nice behavior of consumer
demand functions:
 Convexity: Given any x, y and z in X such that x ≿ y, and y ≿ z, then it
follows that tx + (1-t)y ≿ z, for all 0<t<1.
◦ A set of points is said to be convex if any two points within the set can be joined by a
straight line that contained completely within the set.
 Strict convexity: given x ≠ y and z in X, if x ≿ z, and y ≿ z, then tx + (1-t)y

≿ z, for all 0<t<1. In other words, given any bundle x, its better set is strictly
convex.
 Convexity implies an economic agent prefers averages to extremes. Convex

preferences may have indifference curves that exhibit “flat” while strictly
convex indifference curves have indifference curves that are strictly rotund.
Convexity is a generalization of the neoclassical assumption of “diminishing
marginal rates of substitution”.
40
Preference representation
 There are several equivalent ways of

describing preferences:
◦ Utility function
◦ Indirect utility function
◦ Expenditure function
◦ Etc.
 Inthis chapter we will focus on the Utility
Function approach.
41
Utility
 Given the assumptions of completeness, transitivity,
and continuity, we can formally show that people
are able to rank in order all possible situations from
the least desirable to the most.
 Economists call this ranking utility
 Given that preferences are complete, reflexive,

transitive, continuous and strongly monotonic, then
there exists a continuous utility function which
represents those preferences.
42
Utility
 Economists have abandoned the old-fashioned view of

utility as being a measure of happiness. Instead, the
theory of consumer behavior have been completely
reformulated in terms of consumer preferences, and
utility is seen ONLY as a way to describe preferences:
i.e. More desirable situation offer more utility than less

desirable ones, i.e. if a person prefers situation A to
situation B, we would say that the utility assigned to
option A , denoted U(A), exceeds the utility assigned
to B, U(B).
43
Utility
if A is preferred to B, then the

utility assigned to A exceeds the
utility assigned to B
U(A) > U(B)
44
Utility
 Given preference ordering the set of all
consumption bundles that are indifferent to
each other is called an indifference curve. We
can think of indifference curves as being level
sets of the utility function.
 Nearly any ‘reasonable’ preference can be
depicted by indifference curves. The trick is to
learn what kinds of preferences give rise
to what shapes of indifference curves.
45
Utility
 The Axiom of transitivity implies that
indifference curves cannot intersect.
 The set of all bundles on or above an

indifference curve is called an upper
contour set.
46
Utility
 Economists gradually came to recognize that all that
matters about utility as far as choice behavior was
concerned is whether one bundle has higher utility than
another- how much higher did not really matter.
 Utility measures are not unique, although we can assign
numbers to the utilities. What is important in utility is
the ranking. In other words, the notion of utility is
defined only up to an order preserving (monotonic)
transformation. Utility rankings simply record the
relative desirability of commodity bundles.
47
Utility
 A monotonic transformation of a utility
function is a utility function that
represents the same preference as the
original utility function.
48
Utility
 Utility rankings are ordinal in nature
◦ they record the relative desirability of commodity bundles
 Because utility measures are not unique, it makes
no sense to consider how much more utility is
gained from A than from B
 It make no sense to ask how much more is A
preferred to B, because such questions have no
unique answers.
 It is also impossible to compare utilities between
people
49
Utility
 For example, if one person says his steak
gives him a utility of 5, and another says
he gets a utility of 200, we cannot say that
the latter gets higher utility, just simply
because they may be using different
scales.
50
Utility
 The ceteris paribus assumption:
◦ Since utility refers to overall satisfaction, it clearly is affected by a
number of factors.
◦ A person’s utility is affected not only by his consumption of
physical commodities, but also by psychological attitudes, peer
group pressures, personal experiences, etc.
◦ However, economists usually limit their analysis to quantifiable
options.
◦ Hence, the ceteris paribus assumption is invoked in all economic
analysis of utility-maximizing choices so as to make the analysis
manageable.
◦ Of course, what is held constant and what is allowed to vary will
depend on the particular question being analyzed.
51
Utility
 Arguments of utility functions
◦ The utility function notion is used to indicate how an
individual ranks the particular arguments of the
function being considered. The arguments could be
different types: income, consumption today and
consumption tomorrow, labor-leisure choice, etc.
◦ By changing the arguments of the utility function, we
will be able to focus on specific aspects of an
individual’s decision in a simplified setting.
52
Economic Goods
 In the utility function, the x’s are assumed to
be “goods”
◦ more is preferred to less
Quantity of y
Preferred to x*, y*
y*
?
Worse
than
x*, y* Quantity of x
x*
53
The marginal rate of substitution (MRS)
 Marginal rate of substitution (MRS): A consumer is consuming
two goods i & j.
 Suppose a consumer changes his consumption of good i. MRS
between goods i & j measures by how much the consumer should
change his consumption of good j in order to keep utility constant.
(Example: page 97-98,Varian)
◦ MRS is the negative of the slope of an indifference curve, i.e. The
absolute value of the slope of an indifference curve.
◦ MRS is the rate at which an individual is willing to trade one
good for another while remaining equally well off.
54
◦ MRS is the rate at which an individual is willing to trade one
good for another while remaining equally well off.
◦ Marginal rate of substitution of x for y is equal to the ratio of
the marginal utility of x to the marginal utility of y.
◦ The MRS can be measured by observing a person’s actual
behavior. Hence the ratio of the marginal utilities gives an
observable magnitude- the MRS.
◦ The ratio of the marginal utilities is independent of the particular
form of the utility function chosen.
◦ Sometimes, MRS is measures the marginal willingness to pay.
55
 The MRS does not depend on the utility function
chosen to represent the underlying preferences. (Proof:
Varian, page 98).
 Diminishing MRS: The MRS of x for y, diminishes as x

is progressively substituted for y.
◦ Strict convexity is equivalent to the assumption of
diminishing MRS, and either assumption rules out the
possibility of an indifference curve being straight over
any portion of its length.
56
◦ The assumption of diminishing MRS is equivalent to the
assumption that all combinations of x and y, which are preferred
or indifferent to a particular combination x*, y*, form a convex
set. This definition is equivalent to the assumption that the utility
function is quasi-concave. The assumption of strict quasi-
concavity is used to rule out the possibility of indifference curves
having linear segments.
◦ The assumption of diminishing MRS also mean that the
combination (x1 + x2)/2, (y1 + y2)/2 will be preferred to either
of the initial combinations. Intuitively, this means that “well-
balanced” bundles of commodities are preferred to bundles that
are heavily weighted toward one commodity.
57
Diminishing marginal utility and the
MRS
 Marshal used the assumption of diminishing marginal utility to solve the paradox of
value. Marshal theorized that it is the marginal valuation that an individual places on
a good that determines its value: It is the amount that an individual is willing to pay
for one more pint of water that determines the price of water. Since it might be
taught that this marginal value declines as the quantity of water consumed increases,
Marshal showed why water has low exchange value.
 Intuitively it seems clear that the assumption of the decreasing marginal utility of a
good is related to the assumption of decreasing MRS; both concepts seem to refer
to the same common sense idea of an individual becoming relatively satiated with a
good as more of it is consumed. However, the two concepts are quite different.
 In modern usage, the concept of decreasing MRS has replaced Marshal’s idea
because the discussion of the MRS does not depend so heavily on the utility
concept and appears to be more empirically verifiable statement of the relative
satiation idea.
58
Utility and the MRS
 Suppose an individual’s preferences for
hamburgers (y) and soft drinks (x) can be
represented by
utility  10  x  y
• Solving for y, we get
y = 100/x
• Solving for MRS = -dy/dx:

MRS = -dy/dx = 100/x2
59
Utility and the MRS
MRS = -dy/dx = 100/x2
 Note that as x rises, MRS falls
◦ when x = 5, MRS = 4
◦ when x = 20, MRS = 0.25
60
Marginal Utility
 Suppose that an individual has a utility
function of the form
utility = U(x,y)
 The total differential of U is
U U
dU  dx  dy
x y
• Along any indifference curve, utility is

constant (dU = 0)
61
Deriving the MRS
 Therefore, we get:
U
 x
dy
MRS  
dx Uconstant
U
y
• MRSxy is the ratio of the marginal utility
of x to the marginal utility of y
62
Diminishing Marginal Utility and
the MRS
 Intuitively, it seems that the assumption of
decreasing marginal utility is related to the
concept of a diminishing MRS
63
Convexity of Indifference Curves
 Suppose that the utility function is
utility  x  y
• We can simplify the algebra by taking the
logarithm of this function
U*(x,y) = ln[U(x,y)] = 0.5 ln x + 0.5 ln y
64
 Thus,
U * 0.5
MRS  x  x 
y
U * 0.5 x
y y
65
 If the utility function is
U(x,y) = x + xy + y
 There is no advantage to transforming this
utility function, so
U
x 1 y
MRS  
U 1  x
y
66
 Suppose that the utility function is
utility  x 2  y 2
• For this example, it is easier to use the

transformation
U*(x,y) = [U(x,y)]2 = x2 + y2
67
 Thus,
U *
MRS  x 
2x x

U * 2y y
y
68
Examples of Utility Functions
 Cobb-Douglas Utility
utility = U(x,y) = xy
where  and  are positive constants
◦ The relative sizes of  and  indicate the
relative importance of the goods
69
 Perfect Substitutes
utility = U(x,y) = x + y
Quantity of y
The indifference curves will be linear.
The MRS will be constant along the
indifference curve.
U3
U2
U1
Quantity of x
70
 Perfect Complements
utility = U(x,y) = min (x, y)
Quantity of y
The indifference curves will be
L-shaped. Only by choosing more
of the two goods together can utility
be increased.
U3
U2
U1
Quantity of x
71
 CES Utility (Constant elasticity of
substitution)
utility = U(x,y) = x/ * y/
when   0 and
utility = U(x,y) = ln x + ln y
when  = 0
◦ Perfect substitutes   = 1
◦ Cobb-Douglas   = 0
◦ Perfect complements   = -
72
 CES Utility (Constant elasticity of
substitution)
◦ The elasticity of substitution () is equal to 1/(1
- )
 Perfect substitutes   = 
 Fixed proportions   = 0
73
Homothetic Preferences
 If the MRS depends only on the ratio of the
amounts of the two goods, not on the
quantities of the goods, the utility function
is homothetic
◦ Perfect substitutes  MRS is the same at every
point
◦ Perfect complements  MRS =  if y/x > /,
undefined if y/x = /, and MRS = 0 if y/x < /
74
Homothetic Preferences
 For the general Cobb-Douglas function, the
MRS can be found as
U
 1 
x  x y  y
MRS    
U x y   1
 x
y
75
Nonhomothetic Preferences
 Some utility functions do not exhibit
homothetic preferences
utility = U(x,y) = x + ln y
U
MRS  x 1
 y
U 1
y y
76
The Many-Good Case
 Suppose utility is a function of n goods
given by
utility = U(x1, x2,…, xn)
 The total differential of U is
U U U
dU  dx1  dx2  ...  dx n
x1 x 2 xn
77
The Many-Good Case
 We can find the MRS between any two
goods by setting dU = 0
U U
dU  0  dxi  dx j
xi x j
• Rearranging, we get
U
dx j x i
MRS( x i for x j )   
dx i U
x j
78
Multigood Indifference Surfaces
 We will define an indifference surface as
being the set of points in n dimensions
that satisfy the equation
U(x1,x2,…xn) = k
where k is any preassigned constant
79
Multigood Indifference Surfaces
 If the utility function is quasi-concave, the
set of points for which U  k will be
convex
◦ all of the points on a line joining any two
points on the U = k indifference surface will
also have U  k
80
 If individuals obey certain behavioral
postulates, they will be able to rank all
commodity bundles
◦ the ranking can be represented by a utility
function
◦ in making choices, individuals will act as if they
were maximizing this function
 Utility functions for two goods can be
illustrated by an indifference curve map
81
 The negative of the slope of the indifference
curve measures the marginal rate of
substitution (MRS)
◦ the rate at which an individual would trade an
amount of one good (y) for one more unit of
another good (x)
 MRS decreases as x is substituted for y
◦ individuals prefer some balance in their
consumption choices
82
 A few simple functional forms can capture
important differences in individuals’
preferences for two (or more) goods
◦ Cobb-Douglas function
◦ linear function (perfect substitutes)
◦ fixed proportions function (perfect complements)
◦ CES function
 includes the other three as special cases
83
 It is a simple matter to generalize from two-
good examples to many goods
◦ studying peoples’ choices among many goods
can yield many insights
◦ the mathematics of many goods is not especially
intuitive, so we will rely on two-good cases to
build intuition
84
Chap 2-2
UTILITY MAXIMIZATION
AND CHOICE
85
Consumer behavior
 No real individuals make the kinds of
“lightning calculations” required for utility
maximization
 The utility-maximization model predicts
many aspects of behavior
 Thus, economists assume that people
behave as if they made such calculations
86
Consumer behavior
 The economic model of consumer choice is that consumers
choose the best bundle they can afford, i.e.
◦ Consumers choose the most preferred bundle from their
budget set, or
◦ individuals who are constrained by limited incomes will
behave in such a way as to achieve the highest utility
possible, or
◦ individuals behave as if they maximize utility subject to
budget constraint.
 The utility maximization model predicts behavior even

though no one carries around a computer with his utility
function programmed into it.
87
Consumer behavior
 Utility maximization does not imply “selfish” behavior.

Nothing in the utility maximization model prevents
individuals from deriving satisfaction from philanthropist
decisions or generally from “doing good”.
 In order to maximize utility, given a fixed amount of

money to spend, an individual will buy those quantities of
goods that exhaust his or total income and for which the
psychic rate of trade-off between any two goods (the
MRS) is equal to the rate at which the goods can be traded
one for the other in the market place.
 Nothing in the utility-maximization model prevents

individuals from deriving satisfaction from “doing good”
88
Consumer behavior
 The problem of preference maximization can then be
written as : max u(x),
such that px ≥ m,
where x is a vector of
commodities,
p is a vector of prices and m is money income.
89
Consumer behavior
 Basic features of this problem:
◦ The first issue is whether there will exist a solution to
this problem
 We need to verify that the objective function is
continuous and that the constraint set is closed and
bounded. By assumption, the utility function is
continuous. The constraint set is also certainly
closed and bounded. If pi > 0 for i= 1, …,k and m ≥
0, it is not difficult to show that the constraint set
will be bounded. If some price is zero, the consumer
might want an infinite amount of the corresponding
good. We generally ignore such boundary
conditions.
90
Consumer behavior
◦ The second issue concerns the representation of the
preferences. The maximizing choice x* will be
independent of the choice of utility function used to
represent the preferences.
◦ The third issue is that if we multiply all prices and

income by some positive constant, we will not change
the budget set, and thus we can not change the set of
optimal choices. Roughly speaking, the optimal
choice set is “homogeneous of degree zero” in prices
and income.
91
Consumer behavior
 Under the local non-satiation assumption, a utility-
maximizing bundle x* must meet the budget constraint
with equality. Hence, the consumer’s problem can be
restated as: v(p, m) = max u(x)
such that px = m.
 The function v(p, m) that gives the maximum utility
achievable at given prices and income is called the
indirect utility function. (more on this later)
92
Consumer behavior
 The value of x that solves this problem is the
consumer’s demand bundle. (how much of each
good the consumer desires at a given level of prices
and income).
 The function that relates P and m to the demand
bundle is called the consumer’s demand function.
The demand function is denoted by x(p, m). Strict
convexity of preferences ensures that there is a
unique bundle that maximizes utility.
 The consumer’s demand function is homogeneous
of degree 0 in (p, m).
93
Consumer behavior
 The Lagrangean of the utility maximization problem is
ℐ = u(x) – λ (px-m). The first order condition for
optimization is:
◦ ∂u(x)/ ∂xi – λpi = 0, for all i=1,…,k.
 Dividing the ith first order condition with the jth first
order condition we arrive at
{∂u(x*)/ ∂xi}/ {∂u(x*)/ ∂xj} = pi/pj
 The left hand side fraction shows the MRS between

goods i and j, and the fraction on the right hand side can
be called the economic rate of substitution between
goods i and j.
94
Consumer behavior
 First and second order conditions:
◦ The first order condition is a necessary condition for an
optimum.
◦ Given a diminishing MRS, the first order condition is
both necessary and sufficient condition. Diminishing
MRS implies convex preferences.
95
Consumer behavior
 Meaning of the Lagrangian multiplier:
◦ λ = MU(x1)/p1 = MU(x2)/p2 … = MU(xn)/pn
◦ This equation says that at the utility maximizing point, each
good purchased should yield the same marginal utility per
dollar spent on that good. Each good, therefore, should
have an identical marginal benefit to marginal cost ratio.
The common value for this extra utility is given by the
Lagrangian multiplier of income. Therefore, λ can be
regarded as the marginal utility of an extra dollar of
consumption expenditure, i.e. the marginal utility of
income:
pi = MUi/λ. This equation says that for every good that an
individual buys, the price of that good represents his
evaluation of the utility of his last unit consumed. The price
obviously represents how much the consumer is willing to
pay for that last unit.
96
Consumer behavior
 Corner solutions:
◦ Interior optimum implies positive amount of each goods is
consumed.
◦ Corner solutions imply that zero amount of some good is
consumed.
◦ When corner solutions arise, the optimization conditions must
be modified slightly.
∂ℐ / ∂xi = ∂u/∂xi – λpi ≤ 0 (I = 1, …, k)
And if ∂ℐ / ∂xi = ∂u/∂xi – λpi < 0, then xi = 0. we can rewrite
this condition as pi > {∂u/∂xi}/λ = MU (xi)/λ.
Hence, the optimal condition says that any good whose price
exceeds its marginal value to the consumer will not be
purchased, i.e. individuals will not purchase goods that they
do not believe are worth their money.
97
Examples
 Nicholson, Example 4.2, pages 111-112 and query
 Nicholson, Example 4.3, page 113 and query
 Nicholson, Example 4.4, pages 114-115, and query.
 Varian (1993), Appendix, pages 90-94, Three ways of solving utility
maximization problems and Cobb-Douglas example
 Varian (1993), Cobb-Douglas preferences, page 82-83.
 Varian, (1993), Estimating utility functions, page 83-84.
 Varian (1993), Implications of the MRS condition, page 85-86.
98
Dual to the utility maximization problem
 Many constrained maximization problems have
associated dual constrained-minimization problems. For
utility maximization, the associated dual minimization
problem concerns allocating income in such a way as to
achieve a given utility level with the minimal
expenditure. Note that the goals and constraints have
been reversed from the utility maximization problem.
 Both the primary utility maximization problem and the

dual expenditure minimization problem yield the same
solutions- they are simply alternative ways of viewing
the same process. Often, the expenditure minimization
approach is more useful because expenditures are
directly observable where as utility is not.
99
Indirect Utility
 Because of individual’s desire to maximize utility, given a budget constraint, the

optimal level of utility obtainable will depend indirectly on the prices of the goods
being consumed and on the individual’s income. This dependence is reflected by the
indirect utility function, v. That is, maximum utility = u(x1*, x2*, …, xn*), and since the
x’s are functions of prices and income, then we have v(p1, p2, …, pn, m)
 The indirect utility function v(p, m) gives maximum utility as a function of p and m.
 In both consumer theory and production theory, such indirect approach can be used to
study how changes in economic circumstances affect various kinds of outcomes, such as
utility or firm costs.
Example, Nicholson, page 116, example 4.5.
 Properties of the indirect utility function

◦ v(p, m) is non increasing in p. Similarly, v(p, m) is non-decreasing in m.
◦ v(p, m) is homogeneous of degree 0 in (p, m)
◦ v(p, m), is quasi convex in p; that is, {p: v(p, m) ≤ k) is a convex set for all k.
◦ v(p, m) is continuous at all p » 0, and m>0.
Proves Varian (1992), page 102-103.
100
Expenditure function
 If preferences satisfy the local non-satiation assumption, then v(p,m) will be strictly increasing in m.
Since v(p,m) is strictly increasing in m, we can invert the function and solve for m as a function of
the level of utility, which shows the minimal amount of income necessary to achieve given utility u
at prices p. The function that relates income and utility in this way (i.e. the inverse of the indirect
utility function) is known as the expenditure function and is denoted by e(p, u).
e(p, u) = min px
Such that u(x) ≥ u. The expenditure function gives the minimum cost
of achieving a fixed level of utility. The expenditure function is completely analogous to the
cost function in production theory.
 The expenditure function and the indirect utility function are inverse functions of one another.
 That the expenditure function is an inverse function of the indirect utility function is quite useful in
allowing us to examine the theory of how individuals respond to price changes.
 The expenditure function approach can be useful for addressing practical problems involved in
constructing price indices (more on this later)
 Example: Example 4.6 (Nicholson, page 119-120).
 Price indifference curve: is an indifference curve of all those combinations of prices such that v(p,
m)=k, for some constant k. In other words, price indifference curves are just the level sets of the
indirect utility function. Example: Varian, page 104, figure 7.2.
101
Expenditure function (2)
 Properties of expenditure function

◦ e(p, u) is non-decreasing in p
◦ e(p, u) is homogeneous of degree 1 in P
◦ e(p, u) is concave in p
◦ e(p, u) is continuous in p, for p » 0.
◦ if h(p, u) is the expenditure-minimization bundle
necessary to achieve utility level u, at prices p, then
hi(pi, u) = ∂e(p, u)/∂pi for i=1, …,k, assuming the
derivative exists and that pi >0.
(Proofs, Varian (1992), page 71).
102
Hicksian and Marshalian demand functions
 Hicksian demand function:

◦ The function h(p, u) is called the Hicksian demand function. The
Hicksian demand function is analogous to the conditional factor
demand function in production theory. The Hicksian demand
function tells us what consumption bundle achieves a target level of
utility and minimizes total expenditure. A Hicksian demand
function is also called compensated demand function, a
terminology which comes from viewing the demand function as
being constructed by varying prices and income so as to keep the
consumer at a fixed level of utility. The income changes are
arranged to compensate for the price changes.
 Marshalian demand function:
◦ Hicksian demand functions are not directly observable.
Demand functions expressed as a function of prices and income
are observable. Such demand functions are called Marshalian
demand functions, x(p, m). The Marshallian demand function is
the ordinary demand function discussed in text books.
103
Some important identities
 Some important identities tie together the expenditure function, the indirect
utility function, the Marshalian demand function, and the Hicksian
demand function.
◦ Note that v(p, m*) = max u(x), such that px ≤ m*. Let x* be the solution to
this problem and let u* = u(x*). Consider the expenditure minimization
problem e(p, u) = min px such that u(x) ≥ u*. Under regular conditions, the
answers to these two problems must be the same x*. Hence, the following
four identities:
1. e(p, v(p,m)) ≡ m. The minimum expenditure necessary to reach utility v(p,

m) is m.
2. v(p, e(p, m)) ≡ u. the maximum utility from income e(p,u) is u.
3. xi (p, m) ≡ hi(p, v(p, m). The Marshalian demand at income m is the same
as the Hicksian demand at utility v(p, m)
4. hi(p, u) ≡ xi(p, e(p, u). The Hicksian demand at utility u is the same as the
Marshalian demand at income e(p, u). This identity is perhaps the most
important since it ties the observable Marshalian demand function with the
unobservable Hicksian demand function. Any demand bundle, therefore,
can be expressed either as the solution to the utility maximization problem,
or the expenditure minimization problem.
104
Examples, exercises and extensions
 Example:Varian (1992), page 111-112, The Cobb-

Douglass utility function
 Example:Varian (1992), page 112, The CES utility

function
 Exercises: Nicholson, page 133-124, problems 4.1, 4.3,

4.5, 4.7, 4.9
 Extensions: Nicholson, page 126 E4.1, E4.2, and E4.3
105
The money metric utility function
 Consider some prices p and some given bundles of goods x. We can ask the
following question: how much money would a given consumer need at the
prices p to be as well of as he could by consuming the bundle of goods x?
◦ If we know the consumer’s preferences, we can see how much he needs
to reach the indifference curve passing through x.
◦ The direct money metric utility function gives the minimum
expenditure at prices p necessary to purchase a bundle at least as good
as x.
◦ The money metric utility function occurs so often that it is given this
name. it is also known by names such as the minimum income
function, the direct compensation function and other names.
◦ An alternative definition is: m(p, x) ≡ e(p, u(x))
◦ For fixed x, u(x) is fixed, so m(p, x) behaves exactly like an expenditure
function: monotonic, homogeneous, concave in p, etc.
◦ When p is fixed, m(p, x) is in fact a utility function, because higher
utility requires higher expenditure, the expenditure function is strictly
increasing in u for preferences that exhibit local non-satiation. Hence,
for fixed p, m(p, x) is simply a monotonic transform of the utility
function and is therefore itself a utility function.
106
The money metric utility function (2)
 There is a similar construct for indirect utility, known as the money
metric indirect utility function. It is given by μ (p; q, m) ≡ e(p,
v(q, m). It measures how much money would one need at prices p,
to be as well off as one would be facing prices q and having
income m. Just as in the direct case, the indirect one behaves like an
expenditure function with respect to p, but now it behaves like an
indirect utility function with respect to q and m, since it is simply a
monotonic transformation of an indirect utility function.
 An important feature of the direct and indirect compensation

functions is that they contain only observable arguments. They are
specific direct and indirect utility functions that measure something
of interest.
 Such functions are very useful in welfare economics.
107
Optimization Principle
 To maximize utility, given a fixed amount of income to

spend, an individual will buy the goods and services:
◦ that exhaust his or her total income
◦ for which the psychic rate of trade-off between any goods
(the MRS) is equal to the rate at which goods can be traded
for one another in the marketplace
108
A Numerical Illustration
 Assume that the individual’s MRS = 1

◦ willing to trade one unit of x for one unit of y
 Suppose the price of x = $2 and the price of y = $1
 The individual can be made better off
◦ trade 1 unit of x for 2 units of y in the marketplace
109
The Budget Constraint
 Assume that an individual has I dollars to allocate between
good x and good y
p x x + py y  I
Quantity of y The individual can afford

If all income is spent
I on y, this is the amount to choose only combinations
of y that can be purchased of x and y in the shaded
py
triangle
If all income is spent

on x, this is the amount
of x that can be purchased
Quantity of x
I
px 110
Second-Order Conditions for a Maximum
 The tangency rule is only necessary but not sufficient unless

we assume that MRS is diminishing
◦ if MRS is diminishing, then indifference curves are strictly
convex
 If MRS is not diminishing, then we must check second-order
conditions to ensure that we are at a maximum
111
The n-Good Case
 The individual’s objective is to maximize

utility = U(x1,x2,…,xn)
subject to the budget constraint
I = p1x1 + p2x2 +…+ pnxn
 Set up the Lagrangian:

L = U(x1,x2,…,xn) + (I - p1x1 - p2x2 -…- pnxn)
112
The n-Good Case
 First-order conditions for an interior maximum:

L/x1 = U/x1 - p1 = 0
L/x2 = U/x2 - p2 = 0
•
•
•
L/xn = U/xn - pn = 0
L/ = I - p1x1 - p2x2 - … - pnxn = 0
113
Implications of First-Order Conditions
 For any two goods,
U / xi pi

U / x j p j
• This implies that at the optimal allocation of income
pi
MRS ( xi for x j ) 
pj
114
Interpreting the Lagrangian Multiplier
U / x1 U / x2 U / xn

   ... 
p1 p2 pn
MUx1 MUx2 MUxn

   ... 
p1 p2 pn
  is the marginal utility of an extra dollar of

consumption expenditure
◦ the marginal utility of income
115
Interpreting the Lagrangian Multiplier
 At the margin, the price of a good represents the

consumer’s evaluation of the utility of the last unit
consumed
◦ how much the consumer is willing to pay for the last
unit
MUxi
pi 

116
Corner Solutions
 When corner solutions are involved, the
first-order conditions must be modified:
L/xi = U/xi - pi  0 (i = 1,…,n)
 If L/xi = U/xi - pi < 0, then xi = 0
 This means that
U / xi MUxi
pi  
 
– any good whose price exceeds its marginal value
to the consumer will not be purchased
117
Cobb-Douglas Demand Functions
 Cobb-Douglas utility function:

U(x,y) = xy
 Setting up the Lagrangian:
L = xy + (I - pxx - pyy)
 First-order conditions:
L/x = x-1y - px = 0
L/y = xy-1 - py = 0
L/ = I - pxx - pyy = 0
118
 First-order conditions imply:

y/x = px/py
 Since  +  = 1:
pyy = (/)pxx = [(1- )/]pxx
 Substituting into the budget constraint:

I = pxx + [(1- )/]pxx = (1/)pxx
119
 Solving for x yields

I
x* 
px
• Solving for y yields
I
y* 
py
• The individual will allocate  percent of his
income to good x and  percent of his
income to good y
120
 The Cobb-Douglas utility function is

limited in its ability to explain actual
consumption behavior
◦ the share of income devoted to particular goods
often changes in response to changing economic
conditions
 A more general functional form might be
more useful in explaining consumption
decisions
121
CES Demand
 Assume that  = 0.5

U(x,y) = x0.5 + y0.5
 Setting up the Lagrangian:
L = x0.5 + y0.5 + (I - pxx - pyy)
 First-order conditions:
L/x = 0.5x -0.5 - px = 0
L/y = 0.5y -0.5 - py = 0
L/ = I - pxx - pyy = 0
122
CES Demand
 This means that

(y/x)0.5 = px/py
 Substituting into the budget constraint, we
can solve for the demand functions
I I
x*  y* 
px py
px [1  ] py [1  ]
py px
123
CES Demand
 In these demand functions, the share of

income spent on either x or y is not a
constant
◦ depends on the ratio of the two prices
 The higher is the relative price of x (or y),

the smaller will be the share of income
spent on x (or y)
124
CES Demand
 If  = -1,
U(x,y) = -x -1 - y -1
 First-order conditions imply that
y/x = (px/py)0.5
 The demand functions are
I I
x*  y* 
  py 0. 5
  p 
0 .5


px 1     py 1   x 


  px     py  
 
125
CES Demand
 If  = -,
U(x,y) = Min(x,4y)
 The person will choose only combinations for
which x = 4y
 This means that
I = pxx + pyy = pxx + py(x/4)

I = (px + 0.25py)x
126
CES Demand
 Hence, the demand functions are
I
x* 
px  0.25 py
I
y* 
4 p x  py
127
Indirect Utility Function
 It is often possible to manipulate first-order
conditions to solve for optimal values of
x1,x2,…,xn
 These optimal values will depend on the
prices of all goods and income
x*1 = x1(p1,p2,…,pn,I)
x*2 = x2(p1,p2,…,pn,I)
•
•
•
x*n = xn(p1,p2,…,pn,I)
128
Indirect Utility Function
 We can use the optimal values of the xs to
find the indirect utility function
maximum utility = U(x*1,x*2,…,x*n)
 Substituting for each x*i, we get
maximum utility = V(p1,p2,…,pn,I)
 The optimal level of utility will depend
indirectly on prices and income
◦ if either prices or income were to change, the
maximum possible utility will change
129
The Lump Sum Principle
 Taxes on an individual’s general purchasing

power are superior to taxes on a specific
good
◦ an income tax allows the individual to decide
freely how to allocate remaining income
◦ a tax on a specific good will reduce an
individual’s purchasing power and distort his
choices
130
• A tax on good x would shift the utility-

maximizing choice from point A to point B
Quantity of y
B A
U1
U2
Quantity of x
131
• An income tax that collected the same

amount would shift the budget constraint to I’
Quantity of y Utility is maximized now at point C

I’ on U3
A
B C
U3 U1
U2
Quantity of x
132
Indirect Utility and the Lump Sum Principle
 If the utility function is Cobb-Douglas with  =

 = 0.5, we know that
I I
x*  y* 
2 px 2 py
• So the indirect utility function is
I
V ( px , py , I )  (x*) (y*)
0.5 0.5

2px0.5 py0.5
133
 px=1, py=4, I=8. Indirect utility V=2

 If a tax of £1 was imposed on good x
◦ the individual will purchase x*=2
◦ indirect utility will fall from 2 to 1.41
 An equal-revenue tax will reduce income to £6
◦ indirect utility will fall from 2 to 1.5
134
 If the utility function is fixed proportions with

U = Min(x,4y), we know that
I I
x*  y* 
px  0.25 py 4 p x  py
• So the indirect utility function is
I
V ( px , py , I )  Min( x *,4 y *)  x* 
px  0.25 py
4 I
 4y *  
4 px  py px  0.25 py
135
 If a tax of $1 was imposed on good x

◦ indirect utility will fall from 4 to 8/3
 An equal-revenue tax will reduce income to
$16/3
◦ indirect utility will fall from 4 to 8/3
 Since preferences are rigid, the tax on x does
not distort choices
136
Expenditure Minimization
 Dual minimization problem for utility

maximization
◦ allocating income in such a way as to achieve a
given level of utility with the minimal expenditure
◦ this means that the goal and the constraint have
been reversed
137
Expenditure Minimization
 The individual’s problem is to choose

x1,x2,…,xn to minimize
total expenditures = E = p1x1 + p2x2 +…+ pnxn
subject to the constraint
utility = U1 = U(x1,x2,…,xn)
 The optimal amounts of x1,x2,…,xn will depend
on the prices of the goods and the required
utility level
138
Expenditure Function
 The expenditure function shows the minimal

expenditures necessary to achieve a given
utility level for a particular set of prices
minimal expenditures = E(p1,p2,…,pn,U)
 The expenditure function and the indirect
utility function are inversely related
◦ both depend on market prices but involve different
constraints
139
Two Expenditure Functions
 The indirect utility function in the two-good,

Cobb-Douglas case is
I
V ( p x , py , I ) 
2px0.5 py0.5
• If we interchange the role of utility and

income (expenditure), we will have the
expenditure function
E(px,py,U) = 2px0.5py0.5U
140
Two Expenditure Functions
 For the fixed-proportions case, the indirect

utility function is
I
V ( px , py , I ) 
px  0.25 py
• If we again switch the role of utility and
expenditures, we will have the expenditure
function
E(px,py,U) = (px + 0.25py)U
141
Properties of Expenditure Functions
 Homogeneity
◦ a doubling of all prices will precisely double
the value of required expenditures
 homogeneous of degree one
 Nondecreasing in prices
◦ E/pi  0 for every good, i
 Concave in prices
142
Concavity of Expenditure Function
At p*1, the person spends E(p*1,…)

If he continues to buy the
same set of goods as p*1
E(p1,…) Epseudo changes, his expenditure
function would be Epseudo
E(p1,…)
E(p*1,…) Since his consumption

pattern will likely change,
actual expenditures will be
less than Epseudo such as
E(p1,…)
p*1 p1
143
 To reach a constrained maximum, an

individual should:
◦ spend all available income
◦ choose a commodity bundle such that the MRS
between any two goods is equal to the ratio of
the goods’ prices
 the individual will equate the ratios of the marginal
utility to price for every good that is actually
consumed
144
 Tangency conditions are only first-order

conditions
◦ the individual’s indifference map must exhibit
diminishing MRS
◦ the utility function must be strictly quasi-
concave
145
 Tangency conditions must also be

modified to allow for corner solutions
◦ the ratio of marginal utility to price will be
below the common marginal benefit-marginal
cost ratio for goods actually bought
146
 The individual’s optimal choices implicitly

depend on the parameters of his budget
constraint
◦ choices observed will be implicit functions of
prices and income
◦ utility will also be an indirect function of prices
and income
147
 The dual problem to the constrained utility-

maximization problem is to minimize the
expenditure required to reach a given utility
target
◦ yields the same optimal solution as the primary
problem
◦ leads to expenditure functions in which spending
is a function of the utility target and prices
148
Chap 2-3
Income and Substitution Effect
149
Demand functions
 We can solve the necessary conditions of a utility maximization for the
optimal levels of x1, …, xn (and λ), as a function of all prices and income.
 These can be expressed as n demand functions of the form:

x1* = d1(p1,p2,…,pn,I)
x2* = d2(p1,p2,…,pn,I)
•
•
•
xn* = dn(p1,p2,…,pn,I)
 Once we know the demand functions (d1,…dn), and the values of prices
(P1,P2,…Pn), we can predict the amount of each good by an individual at
given prices and income.
 Individual demand functions are homogeneous of degree zero in all prices

and income.
◦ A function f(x1 , …, xn) is said to be homogeneous of degree k if f(tx1,
…, txn) = tk f(x1 , …, xn).
150
 If there are only two goods (x and y), we can simplify
the notation
x* = x(px,py,I)
y* = y(px,py,I)
 Prices and income are exogenous
◦ the individual has no control over these parameters
151
Homogeneous Demand Functions
 Individual demand functions are homogeneous of degree

zero in all prices and income
 This implies that if we were to double all prices and

income, the optimal quantities demanded will not
change
the budget constraint is unchanged
xi* = di(p1,p2,…,pn,I) = di(tp1,tp2,…,tpn,tI)
152
• With a Cobb-Douglas utility function
utility = U(x,y) = x0.3y0.7
the demand functions are
0 .3 I 0.7I
x*  y* 
px py
• Note that a doubling of both prices and

income would leave x* and y*
unaffected
153
• With a CES utility function
utility = U(x,y) = x0.5 + y0.5
the demand functions are
1 I 1 I
x*   y*  
1  px / py px 1  py / px py
• Note that a doubling of both prices and
income would leave x* and y*
unaffected
154
Income expansion path (income offer curve) and
Engle curves
 Income expansion path: The locus of utility maximizing bundles obtained
as we increase income holding prices fixed.
 Engle curve: a curve derived from the income expansion path and that
relates income to the demand for each commodity (at constant prices). In
other words, the Engle curve is a graph of the demand for one of the goods
as a function of income, with all prices held constant.
 If the income expansion curve (and thus the Engle curve) is a straight line,
through the origin, the consumer is said to have demand curves with unit
income elasticity. That is, such a consumer will consume the same
proportion of each commodity at each level of income.
155
Income expansion path (income offer curve) and Engle
curves (2)
◦ If the income expansion path bends towards one good or the other over
an income range, the consumer consumes more of each good, but
proportionately more of one good (the luxury good), and
proportionately less of the other (the necessary good). The demand for
a luxury good goes up by higher proportion than income over the
income range. The demand for a necessary good, goes up by lesser
proportion than income over the income range.
◦ If the income expansion path bends backwards, (i.e. as income increases

consumer wants to consume less of one of the goods), such good is
called inferior good. Goods for which more income means more
demand are called normal goods.
◦ Whether a good is inferior or not depends on the income level that we

are examining. For example, very poor people may eat more bologna as
their income increases. But after a point the consumption of bologna
would probably decline as income continued to increase. In real life, the
consumption of goods can increase or decrease as income rises.
156
Price offer curve
 Holding income fixed and allowing prices to vary, the budget

line will tilt. The locus of tangencies will yield a curve known
as the price offer curve.
◦ We can have the case where demand increases as prices fall.
Such goods are called Ordinary goods.
◦ We can also have the case where a decrease in prices brings
about a decrease in demand. Such goods are called Giffen
goods. Giffen goods are rare in real-world behaviour.
157
 A good xi for which xi/I  0 over some range of
income is a normal good in that range
 A good xi for which xi/I < 0 over some range of

income is an inferior good in that range
158
Engle’s Law
 Economists have studied the relationship between income and the

consumption of specific goods. Expenditure data are collected from
sample families and are then classified by income levels to see if any
regularities emerge.
 Engle’s law states that the proportion of total expenditure devoted to food
declines as incomes rise. In other words, food is a necessity whose
consumption rises less proportionately than income.
 Cross country comparisons show that people in developing countries spend

a larger percentage of their income on food than individuals in developed
countries.
 This implies that the income elasticity of demand for all nonfood items
must be greater than one

159
Income and substitution effects
 A change in the price of a good alters the slope of the

budget constraint
◦ it also changes the MRS at the consumer’s utility-
maximizing choices
 When the price changes, two effects come into play
◦ substitution effect
◦ income effect
160
 Even if the individual remained on the same indifference
curve when the price changes, his optimal choice will
change because the MRS must equal the new price ratio
the substitution effect
 The price change alters the individual’s “real” income
and therefore he must move to a new indifference curve
the income effect
161
Income and substitution effects
 The effect of a price change on the quantity of a good demanded is

somewhat more complex to analyze than is the effect of a change in
income, because changing a price involves changing not only the position
of the budget constraint but also its slope. Hence, moving to the new utility
maximization choice entails not only moving to another indifference curve
but also changing the MRS.
 Therefore, when price changes, two analytically different effects come into
play: the substitution effect and the income effect.
 The substitution effect implies that even if the consumer were to stay on
the same indifference curve, consumption patterns need to be allocated so
as to equate MRS to the new price ratio.
 The income effect implies that a price change necessarily changes an

individual’s real income so the individual can not stay on the initial
indifference curve but must move on to a new one.
162
Income and substitution effects (2)
 The utility maximization hypothesis implies that for normal goods,

a fall in the price of a good leads to an increase in quantity
purchased because:
(1) the substitution effect causes more to be purchased as the

individual moves along an indifference curve and
(2) the income effect causes more to be purchased because the price
decline has increased the individual’s purchasing power, thereby
permitting movement to a higher indifference curve.
◦ When the price of a normal good rises, similar argument predicts

a decline in the quantity purchased. For inferior goods,
substitution and income effects work in opposite directions, and
so no definite prediction can be made.
163
Income and substitution Effect of a fall in price of X
Quantity of Y
Y**
U2
I/Py U1
Y*
I=P1xX+PyY
I=P2x+PyY
X* XB
X**
Quantity of X
Substitution Income
Effect Effect
Total increase in X 164

Income and substitution Effect of an increase in price of X
Quantity of Y
Y**
U2
I/Py U1
Y*
I=P2xX+PyY
I=P1x+PyY
X** XB
X*
Quantity of X
Income Substitution
Effect Effect
Total increase in X 165

Compensated demand curves
 We can examine reactions to changes in price of a good holding real income

(utility) of the consumer constant.
 Compensated demand curve: A compensated demand curve shows the

relationship between the price of a good and the quantity purchased on the
assumption that other prices and utility are held constant: x* = hx(px, py, u).
 In general, for a normal good, the compensated demand curve is somewhat less
responsive to price changes than the uncompensated curve because the latter
reflects both the substitution and income effects of price changes while the
compensated curve reflects only substitution effects.
 The choice between using compensated or uncompensated demand curves is

economic analysis is largely a matter of convenience. In most empirical work,
uncompensated curves are used since the data on prices and nominal incomes
needed to estimate them are readily available. For some theoretical purposes,
however, compensated demand curves are a more appropriate concept since the
ability to hold utility constant offers a considerable number of advantages. The
concept of consumer surplus is one illustration of these advantages.
 Example: Nicholson, page 146, example 5.3.

166
The Slutsky Equation
 Hicksian, or compensated demand curve, is formally the same as the

conditional factor demands in production theory. Hence, it has all the same
properties/restrictions of the conditional factor demand.
 In the case of the firm, the restrictions imposed on the conditional factor
demands were observable restrictions on firm behavior, because the output
of the firm is an observable variable. In the case of the consumer, this sort
of restrictions does not appear to be of much use since utility is not directly
observable.
 However, even though the compensated demand function is not directly

observable, its derivative can be easily calculated from observable things,
namely the derivative of the Marshalian demand with respect to price and
income. This relationship is known as the Slutsky equation.
167
The Slutsky Equation (2)
 Deriving the Slutsky equation

◦ Assume that there are only two goods (x, y). Let the compensated demand
function be hx(px, py, u).
◦ Expenditure function records the minimum expenditure necessary to attain
a given level of utility. Let this function be minimum expenditure = E(px,
py, u).
◦ Then by definition, hx(px, py, u) = Dx [px, py, E(px, py, u)]. In other words,
the quantity demanded is identical for the compensated and
uncompensated demand functions when the individual’s income is exactly
what is needed to acquire the required utility level.
◦ By partially differentiating the last equation, we get:
∂hx/ ∂px = ∂Dx/ ∂px + (∂Dx/ ∂E* ∂E/∂px). Rearranging terms, we get ∂Dx/
∂px = ∂hx/ ∂px - (∂Dx/∂E * ∂E/∂px).
The first term is the slope of the compensated demand curve, and
measures the change that comes due to change in the price. It is the
substitution effect.
The second term measures the way in which changes in price affects the
demand for x through changes in the necessary expenditure levels. It is
the income effect.
168
The Slutsky Equation (3)
 The substitution effect is

∂hx/ ∂px = ∂x/∂px (u= constant to indicate movement along the
indifference curve)
 The income effect is

- ∂Dx/∂E * ∂E/ ∂px = - ∂x/∂I * x, since changes in income and
expenditure amount to the same thing in the function Dx, and it
is easy to show that ∂E/∂px = x.
 Hence, the utility maximization hypothesis gives the Slutsky

equation as
∂x/∂px = substitution effect + income effect
= ∂x/∂px (U=constant) - x ∂x/∂m
 Example: Nicholson, page 150, Example 5.4 and query.

169
Shephard’s Lemma and Roy’s Identity
 Shephard’s Lemma:
◦ Applying the envelope theorem to the expenditure function yields:
∂E/ ∂px = x(v, px, py)
∂E/ ∂py = y(v, px, py). That is, the compensated demand functions can be
computed directly from the expenditure function by partial
differentiation. This result is called Shephard’s Lemma.
 Roy’s Identity:
◦ Marshalian demand functions can be derived from the indirect utility
function, but the computation is quite complex. Recall that the
Lagrangian expression associated with individual utility maximization is
ℐ = u(x, y) + λ(I – pxx – pyy). Applying the envelope theorem to this
expression yields:
∂u*/ ∂px = ∂ℐ/∂px = - λx and ∂u*/∂I = λ. Hence,
(-∂u*/ ∂px)/ (∂u*/∂I) = (- ∂v/ ∂px)/ (∂v/∂I) = x(px, py, I), where V is the
indirect utility function. This result is called Roy’s identity.
 Extension: Nicholson, page 165, Example E5.3. 170

Relationship among demand concepts
Primal Dual
Maximize u(x, y) Minimize E(x, y)
s.t. I = pxx + pyy s.t. u = (x, y)
Indirect utility Inverses Expenditure

Function Function
U* = V(px, py, I) E* = E(px, py, U)
Roy’s identity
Shephard’s Lemma
Marshalian Compensated
Demand Demand
X = (px, py, I)= X = (px, py, u)=
-(∂V/ ∂px)/ ∂V/ ∂i ∂E/ ∂px
171
Revealed preference
 The principal prediction of the utility maximization model is that the slope
of the compensated demand function (the substitution effect) is negative.
The proof of this assertion relies on the assumption of diminishing MRS,
and the related observation that with a diminishing MRS, the necessary
conditions for utility maximization are also sufficient. Some regard reliance
on hypothesis about an unobservable utility function as a weak foundation
on which to base the theory of demand.
 An alternative approach, which leads to same result, is the theory of

revealed preference, first proposed by Samuelson.
 The theory of revealed preference defines a principle of rationality that is

based on observed behavior and then uses this principle to approximate
an individual’s utility function.
 The principle of Revealed preference: let (x1 , x2) be the chosen

bundle when prices were (p1, p2), and let (y1, y2) be some other bundle
such that p1 x1 + p2 x2 ≥ p1 y1 + p2y2 . Then if the consumer is choosing
the most preferred bundle he can afford, we must have (x1, x2 ) ≻ (y1, y2).
In other words, if the consumer chooses the best bundle, revealed
preference implies preference.
172
Revealed preference (2)
 Weak Axiom of Revealed Preferences (WARP): If (x1, x2) is directly
revealed preferred to (y1, y2), and the two bundles are not the same, then
it cannot happen that (y1, y2) is directly revealed preferred to (x1, x2). In
other words if p1x1 + p2x2 ≥ p1y1 + p2y2, then it must not be the case that
q1y1 + q2y2 ≥ q1x1 + q2x2,
where the pi are prices when the x-bundle was purchased, and the qi are
the prices when the y-bundle was purchased. In other words, this says
that if the y-bundle was affordable when the x-bundle was purchased, then
when the y-bundle was purchased, the x-bundle must not be affordable.
The WARP gives us an observable condition that must be satisfied by all
optimizing agents.
 Strong Axiom of Revealed Preferences (SARP): If (x1, x2) is directly or

indirectly revealed preferred to (y1, y2), and the two bundles are not the
same, then it cannot happen that (y1, y2) is directly or indirectly revealed
preferred to (x1, x2). It is clear that if the observed behavior is optimizing
behavior then it must satisfy the SARP. Any behavior satisfying the SARP
can be thought of as being generated by optimizing behavior, i.e. if the
observed behavior satisfies SARP, we can always find nice well-behaved
preferences that could have generated the observed behavior.
173
Revealed preference (3)
 Most properties that have been developed using
the concept of utility can be proved using SARP.
Therefore, the revealed preference axiom and
the existence of a well-behaved utility functions
are somewhat equivalent conditions. This axiom
provides a believable foundation for utility
theory based on relatively simple comparisons
among alternative budget constraints. These
ideas are important for constructing price
indices and many other applications.
174
Index numbers
 Consider two goods. Let (p1t, p2t ) be prices paid at time t to purchase (x1t,
x2t). In base period b, the prices are (p1b, p2b) and the bundle is (x1b, x2b).
We then have quantity indices (weighted averages of quantities)

Paasche quantity index as:
pq = (p1t x1t + p2t x2t) / (p1t x1b + p2t x2b), and
Laspeyres quantity index as
Lq = (p1b x1t + p2b x2t) / (p1b x1b + p2b x2b),
It turns out that these quantity indices can tell us something interesting
about the consumers welfare.
 If pq >1, we can conclude, using revealed preference that the consumer

must be better off at time t than at time b because he could have
consumed the b consumption bundle in the t situation but chose not to
do so.
175
Index numbers (2)
 If pq <1, it means that the consumer chose the
bundle at time t because the bundle at time b was
not affordable. But, this does not say anything about
the consumer’s ranking of the bundles. Just because
something costs more than you can afford does not
mean that you prefer it to what you are consuming
now.
 The reasoning works in similar way for the Laspeyres

quantity index. If Lq <1, the consumer is better off at
time b than at time t.
176
Index numbers (3)
 Price indices: price index is a weighted average of prices:
◦ Paasche price index:

Pp = (p1tx1t + p2tx2t) / (p1bx1t + p2bx2t)
◦ Laspeyres price index:

Lp = (p1tx1b + p2tx2b) / (p1bx1b + p2bx2b)
Revealed preference does not tell us any thing about welfare

situation of the consumer in periods t and b, because there are
now different prices in the numerator and denominator of the
fractions defining the indices, so the revealed preference
comparison can not be made.
177
Index numbers (4)
 Define a new index of the change in total
expenditure:
M = (p1tx1t + p2tx2t) / (p1bx1b + p2bx2b), that is the ratio of total
expenditure in period t to total expenditure in period b.
Then, if the Paasche price index is greater than M, the bundle

chosen at year b is revealed preferred to the bundle chosen at
year t, that is if the Paasche price index is greater than the
expenditure index, then the consumer must be better off in
period b than in period t. This is quite intuitive. If prices rise by
more than income rises, we should expect that the consumer is
worse off.
Similarly, if the Laspeyres price index is less than M, then the

consumer must be better off in year t than in year b, that is if
prices rise less than income, the consumer must be better off.
178
Substitution effect and consumer
price index
 One practical problem for which the study of
substitution and income effects offer
considerable insight is the construction of price
indexes by which to measure inflation. Common
practice is to choose a market basket of
commodities in a base year and to examine
how the cost of this basket changes over time:
CPI = (∑ipi1xi0)/ (∑ipi0xi0) * 100.
A simple analysis of the expenditure function shows
that CPI computed in this way may overstate the
welfare loss from inflation, since use of fixed market
basket does not permit commodity substitution in
response to change in relative prices.
179
Consumer surplus
 An important problem in applied economics is to develop a
monetary measure of the gains or losses due to price changes.
We need a way to measure the welfare loss that consumers
experience from price changes. In order to make such
calculations, economists have developed the concept of
consumer surplus.
 Recall the expenditure function E(px, py,u). One way to evaluate

the welfare cost of price increase is to compare the
expenditure required to achieve u under these two
circumstances:
Expenditure at px0 = E(px0,py,u) = E0
Expenditure at px1 = E(px1,py,u) = E1. Hence, welfare
change is given by E0 – E1. Therefore, knowledge of the
expenditure function is enough to make the kind of
calculations we need.
180
Consumer surplus (2)
 Recall also that, by Shephard’s Lemma, the
derivative of the expenditure function with
respect to px yields the compensated demand
function, hx i.e, dE(px,py,u)/dpx = hx(px,py,u).
In words, the change in necessary expenditure
brought about by change in px is given by the
quantity of x demanded. To evaluate this change,
in expenditures over price change we must
integrate the last equation. The integral is the
area to the left of the compensated demand
curve (hx) between px0 and px1.
181
Consumer Surplus (3)
 So far our analysis of consumer surplus has made use of
the compensated demand curve. In most actual
applications, the price rise will result in both
substitution and income effects and a loss or gain in
utility to the individual. Therefore, the actual market
reaction to the rise in px, would be a movement on the
Marshalian demand curve. Computations of consumer
surplus using the Marshalian demand curve gives a good
and convenient approximation to welfare changes due
to price changes since information on Marshalian
demand curve is likely to be readily available.
 Example: Nicholson, page 158-159, Example 5.5.
182
Endowments in the budget
constraint
 So far, in our analysis of consumer behavior income was assumed to be exogenous. In more
realistic models of consumer behavior, it may be necessary to consider how income is
generated. The standard way to do this is to think of the consumer as having some
endowment ω = (ω1, ω2, ω3, …, ωk), which can be sold at the current market prices p. The
utility maximization problem becomes:
max u(x)
s.t. px = pω
x
 We can solve this problem using the standard techniques to find demand functions. The net
demand for good I is xi – ωi. The consumer may have negative or positive net demand
functions depending on whether he wants more or less of something than his endowment.
 In this model prices influence the value of what the consumer has to sell as well as what the
consumer wants to buy.
 With normal good, when the price of the good goes up, the substitution effect and the
income effect both push towards reduced consumption. But if the consumer is a net seller of
this good, his actual income increases and this additional endowment income effect may
actually lead to an increase in consumption of the good.
183
Endowments in the budget
constraint (2)
 Labor supply
◦ Suppose a consumer chooses consumption and labor. He also has non-labor
income, m. Let u(c, ℓ) be the utility of consumption and labor. The maximization
problem then becomes:
Max u(c, ℓ)
c, ℓ
s.t. pc = wℓ + m
Let Ĺ be the maximum number of hours that the consumer can work.
Then think of L = Ĺ- ℓ as being leisure.
The utility function for consumption and leisure is u(c, L). The
maximization problem then becomes
Max u(c, L)
c, L
s.t. pc + wL = wĹ + m.
This is essentially the same form we have seen before. Here, the consumer
“sells” her endowment of labor at the price w and then buys some back as
leisure.
184
Endowments in the budget constraint (3)
Labor Supply (cont’d)
Slutsky equation allows us to calculate how the
demand for leisure changes as the wage
rate changes. We have:
dL(p, w, m)/dw =
(∂L(p, w, u)/∂w) + (∂L(p, w, m)/ ∂m) [Ĺ – L]
Note that the term in brackets is nonnegative by

definition. Hence, the derivative of the leisure demand is
the sum of a negative number and a positive
number and is inherently ambiguous in sign. In other
words, an increase in the wage rate can lead to either
an increase or decrease in labor supply.
185
Excercises
 Nicholson, page 162, problems 5.1, 5.3,
5.4, 5.5, 5.6, 5.7, 5.9,
186
Demand relationship among two goods- The
Slutsky Equation
 A change in price of the good affects the demand for a related good in
two ways: the substitution and income effects
∂x/∂py = ∂x/ ∂py (u=constant) - y ∂x/∂m. In the income effect the derivative is
multiplied by the amount of y purchased since that quantity reflects the
extent to which changes in py affect purchasing power.
 The terms on the right hand side have different signs. If indifference
curves are convex, the substitution effect is positive. If we confine to
move along one indifference curve, increases in py increases x and
decreases in py decreases the quantity of x chosen. But, if x is a
normal good, the income effect is clearly negative. Hence, the
combined effect could be either positive or negative.
187
Substitutes and Complements
 For many goods it is easy to generalize the Slutsky
Equation developed for two goods.
◦ ∂xi /∂pj = ∂xi / ∂pj (u=constant) - xj ∂xi /∂m, for any i and j.
This says that the change in the price of any good
induces income and substitution effects that may
change the quantity of every good demanded.
 Two goods are substitutes if one good may, as a

result of changed conditions, replace the other in
use. Two goods are complements if they go
together.
188
Gross Substitute and Gross Complements
 There are two ways of specifying the ideas of substitutes and complements. One focuses on
the gross effects of price changes by including both the income and substitution effects, while
the other looks at substitution effects alone.
 Gross substitutes and Complements:

◦ Two goods xi and xj are said to be gross substitutes if ∂xi /∂pj > 0
◦ Two goods are said to be gross complements if
∂xi /∂pj < 0.
 That is, two goods are gross substitutes if the rise in the price of one good causes more of
the other good to be bought. They are gross complements if a rise in the price of one good
causes less of the other good to be purchased.
 It might always be reasonable to think of gross substitutes and complements since the
substitution and income effects are always combined.
 However, there are several things that are undesirable about the concepts of gross
substitutes and complements. The most important problem is that these definitions are not
symmetric. It is possible, by the definitions, for x1 to be a substitute for x2 and at the same
time for x2 to be a complement for x1. the presence of income effect can produce
paradoxical results. Example food and yo-yo.
189
Net Substitutes and Net Complements
 Because of the ambiguities in the definition of gross substitutes and complements,
an alternative definition that focuses only on substitution effects is some times
used
 Net substitutes and Net complements: xi and xj are said to be net substitutes if
∂xi /∂pj (u=constant) > 0. two goods are net complements if ∂xi /∂pj
(u=constant) < 0.
 This definition is both intuitively appealing (because it looks only at the

shape of an indifference curve) and theoretically desirable (because it is
unambiguous).
 The definitions of net substitutes and net complements are perfectly

symmetric, i.e ∂xi /∂pj (u=constant) = ∂xj /∂pi (u=constant). The substitution
effect of a change in pi on good xj is identical to the substitution effect of a
change in pj on the quantity of xi chosen.
 In a world of only two goods, they must be net substitutes although they
may either be gross substitutes or gross substitutes or gross complements.
In fact, there can only be few complementary relationships in the net sense.
Net substitution is the prevalent relationship among goods.
190
Composite commodities
 If an individual consumes n goods, his demand functions will
reflect n(n+1)/2 different substitution effects. When n is very
large (as is the case for all the specific goods individuals
actually consume) this general case can be unmanageable. It is
often far more convenient to group goods into larger
aggregates such as food clothing, shelter and so on. At the
most extreme case of aggregates, we might wish to examine
one specific good and its relationship to all other goods.
 Composite commodity: A composite commodity is a group of

goods for which all prices move together. These goods can be
treated as a single commodity in that the individual; behaves
as if he or she were choosing between other goods and total
spending on the entire composite group.
191
Composite commodities (2)
 Composite commodity theorem: Suppose a consumer is consuming n commodities, but we
interested only in one of them x1. In general, the demand for x1 will depend on the individual
prices of the other n-1 commodities. But, if all these prices move together, it may make sense
to lump them together into a single composite commodity, y. That is if we let p2, …, pn
represent the initial prices of these goods, then we assume that these prices will move
together. They might all double, decline by 50%, but the relative prices would not change. Now,
we define the composite commodity y to be the total expenditure on x1 … x2,, using the initial
prices: y = p20 + … + pn0. This person’s budget constraint is given by m = p1x1 + y. Assume all
the prices in the composite commodity change by a factor of t>0. Now the budget constrain is
m = p1x1 + ty. Hence the factor of proportionality, t, plays the same role in the person’s budget
constraint as did the price of y (py) in the two-goods model. Changes in p1 or in t induce the
same kind of substitution effects we have been studying. So long as p2, …, pn move together, we
can therefore confine our examination of demand to the choice between x1 and everything
else.
 The composite commodity theorem can be shown to apply to any group of commodities
whose relative prices move together. It is possible to have more than one composite
commodity if there are several groupings that obey the theorem (eg. food, clothing, shelter,
etc.)
192
Applications, Exercises and
Extensions
 Applications: Nicholson, page 185-86
 Exercises: Nicholson, pages 186-187,
problems 6.1, 6.3, 6.5, 6.6, 6.7
 Extensions: Nicholson, pages 187-188,

E6.1, E6.2, E6.3, E6.4 and E6.5 and E6.6
193
Market demand functions and curves
 The total demand of a commodity x is simply the sum of the amounts demanded by
individuals. The market demand will obviously depend on the parameters of the
individual demand functions (Pi‘s and m’s). In the two-goods two-individuals case, x =
x1 + x2 = Dx1 (px ,py ,m1) + Dx2 (px ,py ,m2)
 The market demand function for a particular good xi is the sum of each individual’s
demand function for that good:
xi = ∑ j=1m xij = xi (p1, … pn, m1, …, mm).
 The demand for xi depends on the entire distribution of individual’s incomes.

Although some analysts refer to the changes in aggregate total purchasing power on
the demand for a good, this approach may be misleading simplification, since the
actual effect of such a change on total demand will depend on how the income
changes are distributed among individuals.
 The market demand curve for xi is constructed from the demand function by varying
pi while holding all other variables constant.
 The market demand curve is simply the horizontal sum of each individual demand
curve.
194
Shifts in the market demand curve
 Change in demand and change in quantity demanded:
◦ Change in demand means a shift in the demand curve- arises due to changes I
income and prices of other goods
◦ Change in quantity demanded means movement along the demand curve- arises
due to change in own price.
 Analysis of why market demand curves shift requires analysis of why

individual demand curves shift
 The effect of changes in total income on market demand will depend
greatly on how income changes are distributed among individuals. For
example an income tax change that favored low-income consumers might
have a substantial effect on the demand for food and general retail
purchases but little on luxury goods.
 If py rises and x and y are regarded by most people as being gross
substitutes, the market demand for x will rise. If most people regard the
two goods as gross complements, an in increase in py will cause the
market demand for x shift inwards.
195
Price Elasticity of demand
 Price elasticity of demand eq,p = percentage change
in q/percentage change in p = ∂q/ ∂p * p/q. This
elasticity records how q changes in percentage
terms in response to a percentage change in p.
eq,p is usually negative, except in the Giffen good
case.
 eq,p < -1  elastic demand
eq,p > -1  inelastic demand
eq,p = 1 unitary elastic demand
 Usually elasticities are expressed in their absolute
values. Hence elasticity grater than 1  elastic
demand; elasticity = 1  unitary elatisticity; elasticity
<1  inelastic demand.
196
Price elasticity and total
expenditure
 Total expenditure = pq
 Differentiating pq wit respect to p (since q is a function of p) yields:
∂pq(p)/ ∂p = q + p ∂q/ ∂p. Dividing both sides by q, we get
(∂pq/∂p)/q = 1 + ∂q/∂p . p/q = 1 + eq,p.
 Since q is positive, the sign of ∂pq/∂p will depend on whether eq,p is

greater than or less than -1.
 If eq,p > -1, demand is inelastic and price and total expenditure move in the
same direction. Such situation has been observed in the market for
agricultural products. Demand for food is price inelastic, an increase in its
price actually increases total expenditure on food.
 If eq,p < -1, demand is elastic and price and total expenditure will move in
opposite direction.
197
Income elasticity of demand
 eq,m = percentage change in quantity
demanded/percentage change in income = ∂q/ ∂m *
m/q.
 For a normal good eq,m is positive, since ∂q/ ∂m is
positive. For an inferior good, eq,m is negative.
 Among normal goods, it is important to know
whether eq,m is greater than or less than one. Goods
for which eq,m > 1 might be called luxury good. Goods
for which eq,m < might be called necessities.
198
Cross-price elasticity
 eq,p’ = ∂q/∂p’
* p’/q. If q and other goods
are gross substitutes, ∂q/∂p’ will be
positive as will be eq,p’. When the goods
are gross complements, ∂q/∂p’ will be
negative, as will be eq,p’.
199
Sum of elasticities of all goods
 Recall that the budget constraint for two goods is pxx + pyy = m.
Differentiating the budget constraint with respect to m, we have px ∂x/∂m
+ py ∂y/∂m = 1. Multiplying eah term by a form of 1, 
(pxx)/m * ∂x/∂m* m/x + (pyy)/m * ∂y/∂m *m/y = 1.
 (pxx)/m is the proportion of income spent on good x, and (p yy)/m is
the proportion of income spent on y. Using sx to denote the
proportion of income spent on x and sy on y, we have sx ex,m + syey,m
= 1, an equation sometimes referred to as generalized Engle’s law.
That is, the weighted sum of the income elasticity's of demand for all
goods must be unity. For example, when income increases by 10%,
the budget constraint requires that purchases as a whole increase
by 10%.
 It shows that for every good (or group of goods) that has an income
elasticity of income less than one, there must be goods that have
income elasticities greater than one. If there are only two goods,
knowledge of the one good’s income elasticity and of the share of
income devoted to that good permits calculation of the income
elasticity of the other good.
200
Slutsky equation in elasticities
 Recall that the Slutsky equation was given as:
∂x/∂px = ∂x/∂px (u=constant) - x ∂x/∂m. Multiplying this equation by
px/x, we get, ∂x/∂px * px/x = ∂x/∂px * px/x (u=constant) – pxx* ∂x/∂m *
1/x. Multiplying the numerator and denominator of the final term by
m, we have ∂x/∂px * px/x = ∂x/∂px * px/x (u=constant) - pxx*/m * ∂x/∂m
*m/x.
 The substitution elasticity:

◦ esx,px = ∂x/∂px * px/x (u=constant), which shows how the compensated
demand for x responds to proportional compensated price changes.
 Hence, ex,px = esx,px – sxex,i. This is the Slutsky relationship in

elasticity form. It shows the price elasticity of demand can be
disaggregated into substitution and income components and that the
relative size of the income component depends on the proportion of
total expenditure devoted to the good in question (that is, on s x).
201
Exercises, applications and extensions
 Nicholson, pages 203—204, Types of demand

curves
 Application, Nicholson, page 208-212,
 Problems, Nicholson, page 212-213, problems
7.1, 7.3, 7.4, 7.5, 7.6, 7.7, 7.9.
202
Chapter 3:
Theory of Production
203
Firms
 Firms coordinate the transformation of inputs
into outputs. In other words, firms are
organizations that use inputs to produce
economic goods.
 Firms may be large organizations (such as car

factories, IBM, large commercial farms, etc) or
small ones such as department stores, or self-
employed individuals.
204
Measurement of inputs and outputs
 The simplest and most common way to describe the technology of the
firm is the production function. However, there are other ways to
describe firm technologies that are both more general and more useful in
certain settings. We will focus on the concept of production function in
this chapter, but will present the different ways of describing technology
first.
 Measurement of inputs and outputs

◦ It is usually most useful to think of inputs and outputs as being measured in
terms of flows, i.e. certain amount per given period. One should explicitly
include a time dimension in the specification of inputs and outputs, to avoid
using incommensurate units or confuse stocks and flows. For example, if we
measure labor time in hours per week, we should measure capital services in
hours per week, and output in units per week.
◦ Sometimes, we may want to distinguish inputs and outputs by the calendar
time in which they are available, the location in which they are available, and
even the circumstances in which they are available. This will enable us to
capture some aspects of temporal or spatial nature of production.
◦ The level of detail that we will use in specifying inputs and outputs will depend
on the problem at hand.
205
Specification of technology
 Suppose that the firm has n possible goods to serve as inputs
and/or outputs. If a firm uses yji units of a good j as an input
and produces yj0 of the good as an output, then the net
output of good j is given by yj = yj0 – yji. If the net output of
good j is positive, then the firm is producing more of the good
j than it uses as an input; if the net output is negative, then
the firm is using more of good j than it produces.
 A production plan is simply a list of net outputs of various

goods. We can represent a production plan by a vector y in
Rn, where yj is negative if it is used as net input, and positive if
it serves as net output.
 Production possibility set is the set of technologically

feasible production plans, and can be denoted by Y, a subset
of Rn. This set gives a complete description of the
technological possibilities facing the firm.
206
Specification of technology (2)
 Input requirement set: The set of all input bundles that produce at
least y units of output :V(y) = {x in R+n: (y, -x) is in Y}.
 Isoquant: The isoquant gives all input bundles that produce exactly y
units of output: Q(y) = {x in R+n: x is in V(y) and x is not in V(y’) for y’ > y}.
(More on this later).
 Production function: If a firm has only one output, f(x) = {y in R: y is the

maximum output associated with –x in Y}.
 Transformation function: There is an n-dimensional analog of a

production function that will be useful in the study of general equilibrium
theory. A production plan y is technologically efficient if there is no y’ in Y
such that y’ > y. In other words, a production plan is efficient if there is no
way to produce more output with the same inputs or to produce the
same output with less inputs. Just as the production function picks out the
maximum scalar output as a function of the inputs, the transformation
function picks out the maximal vectors of net outputs.
207
Specification of technology (3)
 Examples:
◦ Cobb-Douglas technology:
Y = {(y, -x1, -x2) in R3 : y ≤ x1ax21-a}
V(y) = {(x1 , x2) in R+2 : y ≤ x1ax21-a}
Q(y) = {(x1 , x2) in R+2 : y = x1ax21-a}
Y(z)= {(y, -x1, -x2) in R3 : y ≤ x1ax21-a, x2=z}
T(y, x1,x2) = y- x1ax21-a
F (x1,x2) = x1ax21-a
208
Properties of technologies
 Monotonic technologies
◦ If x is in V(y) and x’ ≥ x, then x’ is in V(y). Or
◦ If y is in Y and y’ ≤ y, then y’ must be in Y.
 Convex technologies:
◦ If x and x’ are in V(y), then tx + (1-t)x’ is in V(y) for all 0 ≤ t ≤ 1.
That is V(y) is a convex set. Convexity of production technology
says that if two input bundles can each produce y units of output,
then any weighted average of the bundles can also produce y
units of output.
◦ From the production set side, convexity says that if y and y’ are
in Y, then ty + (1-t)y’ is also in Y, for 0 ≤ t ≤ 1
◦ Convex production set implies convex input requirement set
◦ Convex input requirement set is equivalent to quasiconcave
production function.
209
Properties of technologies (2)
 Regular technologies:
◦ V(y) is a closed and non-empty set for all y ≥
0.
◦ The assumption that V(y) is nonempty
requires that there is some conceivable way
to produce any given level of output.
210
Production Functions
 The firms production function for a particular good q, q = f( K, L),
shows the maximum amount of the good that can be produced using
alternative combinations of capital and labor.
 Consider q = f(K, L, M, …), where k represents capital (machine) usage
during the period, L represents hours of labor input, M represents raw
materials use, etc.
 This function might represent a farmer’s output of wheat during one year,
as a function of the quantity of capital equipment employed during the
year, the amount of labor used, the amount of land under cultivation, the
amount of fertilizer applied, the amount of seed used etc. The function
records the fact that there are many different ways of producing a given
amount of wheat. For example, the farmer could use labor intensive
technique with only small amount of mechanical equipment, or capital
intensive technique with small amount of labor.
 The important question from economics point of view is how the levels of
q, k, l and m are chosen by the firm
211
Marginal physical product
 The Marginal physical product of an input
is the additional output that can be
produced by employing one more unit of
that input while holding all other inputs
constant.
◦ MPPk = ∂q/∂k
◦ MPPL= ∂q/∂L
212
Diminishing Marginal Productivity
 Diminishing Marginal Productivity
◦ Generally, the total product (TP) curve for an input will embody the
assumption that MP will increase rapidly initially, increases less rapidly
next, reaches zero, and then becomes negative. (Graph: Nicholson,
page 296).
◦ The TP curve shows the assumption that an input’s marginal physical
productivity eventually declines as more of the input is added while all
other inputs are held constant.
◦ Mathematically, the assumption of diminishing marginal productivity is an
assumption about the second order partial derivatives of the
production function:
 ∂MPK/ ∂K = ∂2q/ ∂K2 = fKK < 0.
 ∂MPL/ ∂L = ∂2q/ ∂L2 = fLL < 0.
◦ The Diminishing MP was the basis for Malthus to predict that
population growth would outstrip food production, because
population is increasing but land resource is limited and so
marginal productivity of labor would decline.
213
Diminishing MP (2)
 Consider one input, labor. The marginal physical
productivity of labor (MPL) is the slope of the total
productivity curve. MPL reaches a maximum at L*,
beyond which it declines as more labor input is used.
This is a reflection of the assumption of diminishing
marginal product. MPL equals zero at some level of
labor use, L*** at which point TPL reaches maximum.
Beyond L***, further additions of labor input actually
reduces output. The point L*** is sometimes called
the intensive margin of production. Production
will not take place beyond L***. (Figure 11.1,
Nicholson, page 296).
214
Average Physical productivity
 Although the concept of average productivity (AP) is not nearly as
important in theoretical economic discussions as marginal productivity, it
receives a great deal of attention in empirical discussions.
◦ APL = q/L; APK = q/K
◦ Geometrically, APL is the slope of the chord drawn from the origin to the
relevant point on the TPL curve.
◦ MPL and APL will be equal at some level of L, L**. At this point, the chord
through the origin is tangent to the TPL. Hence MPL = APL. At this point, the APL
is at its maximum. For levels of labor input less than L**, MPL > APL. For labor
inputs greater than L**, APL > MPL.
◦ The point L** is sometimes called the extensive margin of production.
◦ Only for levels of labor usage greater than L** will output per labor hour (APL)
exceed what an extra worker is able to produce (hence what the worker might
be paid). Hence production is profitable only beyond L**.
Example: Nicholson, page 298, Example 11.1.
215
Isoquants and the marginal rate of
technical substitution (MRTS)
 One way to describe a whole production
function is by using its isoquant map:
◦ An isoquant shows those combinations of K and L
that can produce a given level of output.
Mathematically, an isoquant records the set of inputs
that satisfy f(K, L) = q0. Isoquants record successively
higher levels of output as we move in northeasterly
direction.
◦ Isoquants and indifference curves are similar
concepts, but isoquants represent measurable
quantities. Hence, we should be interested in studying
the shape of the isoquant curves and their related
production functions more than than we examine the
exact shape of the utility functions.
216
Isoquants and the MRTS
 The slope of an isoquant shows how one input can be
substituted (traded) for another while holding output
constant. Examination of the slope of an isoquant will give us
some information about the technical possibility of
substituting one input for another.
 Defn: for two inputs L and K, the marginal rate of technical
substitution (RTS) of labor for capital shows the rate at which
labor can be substituted for capital while holding output
constant.
◦ RTS (L for K) = -dK/dL (q=q0).
◦ The particular value of the RTS depends not only on the level of output
but also on the quantities of capital and labor being used. In other
words, the Value of the RTS depends on the point on the isoquant map
at which the slope is to be measured.
217
RTS and Marginal Productivities
 The RTS (of L for K) is equal to the ratio
of the marginal physical productivity of
labor (MPL) to the marginal physical
productivity of capital (MPK). Proof:
◦ Along an isoquant, dq = ∂f/∂L . dL + ∂f/∂K .
dK = MPL dL + MPK dK = 0  -dK/dL =
RTS (L for K) = MPL/MPK.
218
Law of diminishing RTS
 For convex isoquants, the RTS is diminishing. It is usually
assumed that for high ratios of K to L, the RTS is a large
positive number indicating that a great deal of capital
can be given up if one more unit of labor is to be
substituted. On the other hand when a lot of labor is
already being used, the RTS is low, signifying that only a
small amount of capital can be traded for an additional
unit of labor.
 It is generally not possible to derive diminishing RTS
from the assumption of diminishing marginal
productivity alone.
 Example: Nicholson, page 303, Example 11.2.
219
Cross-productivity effects
 It seems reasonable the the cross-partial derivatives of
the MPs, fKL = fLK should be positive. Since fLK =
∂MPL/∂K, we are interested in how an increase in
capital affects the marginal physical productivity of
labor (or vice versa). For example, it seems
reasonable that when workers have more capital,
they would have higher marginal productivity. But,
although this is probably the most prevalent case, it
does not necessarily have to be so. Many plausible
production functions have fLK <0, at least for some
input values.
 When we assume diminishing RTS, we are making

somewhat stronger assumption than simply
diminishing marginal productivities for each input.
220
Returns to Scale
 There are two important questions to be asked about production functions:
◦ The first important question is how output responds to increase in all inputs together
 the question of returns to scale.
◦ The second important question is how easy it is to substitute one input for the other 
the issue of elasticity of substitution
 Returns to scale: Given q = f(K, L), and all inputs are multiplied by the same
positive constant, m (m>1) we have:
◦ f(mK, mL) = mf(K, L) = mq  Constant returms to scale
◦ f(mK, mL) < mf(K, L) = mq  Decreasing returns to scale
◦ f(mK, mL) > mf(K, L) = mq  Increasing returns to scale
 It is theoretically possible for a function to exhibit constant returns to scale for

some levels of input usage and increasing or decreasing returns for other levels.
Often, economists refer to the degree of returns to scale of a production function
with the implicit notion that only a fairly narrow range of variation in input usage
and the related level of outputs are being considered.
221
Returns to Scale (2)
The n-Input case
 The definition of returns to scale can be easily generalized to a production
function with n inputs: Given q = f(x1, x2, …, xn), and all inputs are multiplied by m,
we have f(mx1, mx2, …, mxn) = mk f(x1, x2, …, xn)= mkq for some constant k. If k=1,
we have constant returns to scale. K<1 implies decreasing returns to scale, and
k>1 implies increasing returns to scale.
 The crucial part of this mathematical definition is the requirement that all inputs
be increased b y the same proportion. In the real world production process, this
provision may make little economic sense. For example, a firm may have only one
boss and that number would not necessarily be doubled even if all inputs were
doubled. Or the output of a farm may depend on soil fertility. It may not literally be
possible to double the acres planted while maintaining fertility, since the new land
may not as good as that already under cultivation. Hence, some inputs may have to
be fixed. (or at least imperfectly variable). In such cases some degree of diminishing
productivity seems likely, although this cannot be properly called diminishing
returns to scale, because of the presence of quasi-fixed inputs.
222
Constant returns to scale and the RTS
 Constant returns to scale production functions occupy an important place in
economic theory, because there are economic reasons for expecting an industry’s
production function to exhibit constant returns.
 If all production in an industry is carried out in plants of an ‘efficient’ size, then

doubling all inputs could most reasonably be accomplished by doubling the number
of these plants. But presumably this would double output since there are now
exactly twice as many plants. Hence, the industry would behave as if it had a
constant returns to scale production function. As long as doubling of inputs is
brought about by doubling the number of optimally sized plants, this will be the
case.
 Constant returns to scale production functions also have the interesting

theoretical property that the RTS between two factors depends only on the ratio
of the two factors, not on the scale of production. Geometrically, all the isoquants
of a constant returns to scale production function are radial blow-ups of the unit
isoquant. (figure 11.3, Nicholson, page 307). Along any ray through the origin,
(a line along which K/L is constant) the slope of the isoquants is the same.
 Isoquants of a constant returns to scale technology are equally spaced as output

expands, thus exhibiting the the constant proportional relationship between
increases in all. Inputs and increases in output.
223
The Elasticity of Substitution
 The second important characteristic of production functions is
how easy it is to substitute one input for another. This is the
question of a single isoquant rather than a question of a whole
isoquant map.
 If RTS does not change at all for changes in K/L, we might say that
substitution is easy, since the ratio of the marginal productivities of
the two inputs does not change as the input mix changes. If the RTS
changes rapidly for small changes in K/L, we would say that the
substitution is difficult, since minor variations in the input mix will
have substantial effect on the inputs’ relative productivities. A scale
free measurement of this responsiveness is provided by the
Elasticity of Substitution.
 For the production function q = f(K, L), the elasticity of substitution
(σ) measures the proiportionate change in K/L relative to the
proportionate change in RTS along an isoquant.
◦ σ = percent Δ(K/L)/Percent ΔRTS = d(K/L)/dRTS * RTS/(K/L).
224
The Elasticity of Substitution (2)
 Since along an isoquant, K/L and RTS are assumed to move in the same direction,
the value of σ is always positive. (Figure 11.4, page 308, Nicholson).
 If σ is high, the RTS will not change much relative to K/L, and the isoquant will be
relatively flat. On the other hand, if σ is low, the isoquant is sharply curved (the
RTS will change by a substantial amount as K/L changes).
 In general, it is possible that the elasticity of substitution will vary as one moves
along an isoquant and as the scale of production changes. Frequently, however, it is
convenient to assume that σ is constant along an isoquant.
 If constant returns to scale are assumed, then, since all the isoquants are merely
radial blowups of each other, σ will be the same along all isoquants. Many
investigations of real-world production functions have centered on this constant
returns to scale, constant elasticity of substitution type.
 Examples of production functions where σ is constant include linear production

function (σ = infinity), fixed proportions (Leontef function) (σ =0), Cobb-Douglas
production function (σ =1) and Constant elasticitry of Substitution (CES)
production function (σ = some constant). (more on this later)
225
The Elasticity of Substitution (3)
 The n-Input case
◦ Generalization of the elasticity of substitution to the many input case raises several
complications.
◦ One can adopt the same definition of σ used in the two-input case, but add the
requirement that all other inputs are held constant. This requirement (which is not
relevant when we consider only two inputs) restricts the value this potential definition. In
real-world production processes, it is likely that any change in the ratio of two inputs will
also be accompanied by changes in the levels of other inputs. Some of these inputs may
be complementary with the ones being changed while others may be substitutes, and to
hold them constant creates a rather artificial restriction. For this reason, an alternative
definition of the elasticity of substitution that permits such complementarity or
substitutability is generally used in the n-good cases (more on this later).
 Examples: Nicholson, pp 309-314, and example 11.3, page 314-315)
226
Technical progress
 Methods of production improve over time, and
it is important to be able to capture these
improvements in the production function
concept.
 (Figure 11.7, Nicholson, page 316).
 Productivity of an input may change due to
technical progress or increase in the use of
other inputs (if complementary). Use of the
production function concept can help to
differentiate between these two concepts.
227
Measuring Technical progress
 One observation about technical progress is that historically the rate of growth of output over time has exceeded the
growth rate that can be attributed to the growth in inputs.
 Suppose q = A(t)f(K, L) is the production function for some good or for societies output as a whole. The term A(t)
represents all the influences that go into determining q besides the inputs. Changes in A overtime represent technical
progress. For this reason A is shown as a function of time.we expect dA/dt >0, i.e. particular levels of the inputs become
more productive over time.
 Differentiating the production function with respect to time, we get:

dq/dt = dA/dt*f(K, L) + A*(df(k,L)/dt) =
dA/dt*q/A + (q/f(K, L) * [∂f/ ∂k*dK/dt + ∂f/ ∂L*dL/dt] .
dividing by q gives,
(dq/dt)/q = (dA/dt)/A + (∂f/ ∂k)/f(K, L)* dK/dt + (∂f/ ∂L)/f(K, L) *dL/dt OR
(dq/dt)/q = (dA/dt)/A + (∂f/ ∂k)* K/f(K, L)* (dK/dt)/K + (∂f/ ∂L)* L/f(K, L) *(dL/dt)/L
For any variable x, (dx/dt)/x is the rate of growth of x per unit of time. If we denote
this by Gx, our last equation becomes:
Gq = GA + (∂f/ ∂k)* K/f(K, L)* GK + (∂f/ ∂L)* L/f(K, L) * GL. We know that
(∂f/ ∂k)* K/f(K, L)* is the elasticity of output with respect to K (eqK), and (∂f/ ∂L)* L/f(K,
L) is the elasticity of output with respect to L (eqL).
Therefore, the growth equation becomes:

Gq = GA + eqKGK + eqLGL 228
Measuring Technical progress (2)
 This shows that the rate of growth of output can be broken down
into the sum of two components: growth attributed to changes in
inputs (K and L), and other residual growth (that is changes in A)
that represents technical progress.
 Example: Solow’s study regarding the entire US economy between
1909 and 1949, recorded the following values for the terms in the
equation:
Gq = 2.75 % per year
GL = 1.00 % per year
GK = 1.75 % per year
eq,L= 0.65
eq,K = 0.35
Hence, GA = Gq - eq,L GL - eq,KGK = 1.50. solow concluded that technology
advanced at a rate of 1.5% per year during 1909 -1949. More than one-
half of the growth in real output could be attributed to technical change
rather than to growth in the physical quantities of the inputs.
229
Types of technical progress
 An important empirical question is “in what precise way does the
technical change factor (A(t)) enter into the production function. There
are three possible ways in which this might happen:
1. Neutral technical progress: q = A(t)f(K, L). That is, technical progress affects
inputs equally.
2. Capital augmenting technical progress: q = f(A(t)K, L). That is, technical
progress affects only capital. Machine-hours become more productive over
time as new technology is applied.
3. Labor augmenting technical progress: q = f(K, A(t)L). That is, technical
progress affects only the quality of labor in the production function.
– Each of these types of technical progress has the effect of shifting the
production function. Overtime, more output can be obtained from given
amount of inputs. One important role of empirical investigation is to
identify the relative role importance of each type of technical progress.
– Example: Nicholson, page 320, Example 11.4.
230
Applications, Extensions and
Exercises
 Applications: Nicholson, page 321-323,
 Extensions: Nicholson, page pages 325-
326
 Exercises: Nicholson, page 323-
325.problems 11.2, 11.3, 11.5, 11.6, 11.7,
11.11.
231
Profit Maximization and Supply
232
Economic Profits
 Economic profit is defined to be the difference between
the revenue a firm receives and the costs that it incurs.
All costs must be included in the calculation of profit.
 Both costs and revenues depend on the actions of the

firm. These actions may take many forms. We can write
revenue as a function of the level of operations of some
n actions, R(a1, …, an), and costs as a function of these
same n activities, C(a1, …, an).
 A basic assumption of most economic analysis of firm

behavior is that a firm acts so as to maximize its profits.
That is, a firms chooses actions (a1, …, an) so as to
maximize R(a1, …, an) - C(a1, …, an).
233
Profit Maximization
 A profit maximizing firm chooses both its inputs and outputs with the sole
goal of achieving maximum economic profits. That is, the firm seeks to
make the difference between its total revenue and its total economic cost
as large as possible.
 The profit maximization assumption has a lot to recommend both in

terms of its ability to yield interesting theoretical results, and to ex-plain
actual firm’s decisions.
 If firms are profit maximizers, they will make decisions in a “marginal” way.
The entrepreneur will perform the conceptual experiment of adjusting
those variables that can be controlled until it is impossible to increase
profits further, which involves looking at the incremental (marginal) profit
obtainable from producing one more unit of output.
234
Profit Maximization (2)
 Define Total revenue as TR(q) = p(q).q, and total cost as
TC(q). Profit then is π (q) = p(q).q - TC(q). The
necessary condition for choosing the value of q that
maximizes profits is dπ/dq = dTR/dq – dTC/dq = 0 
dTR/dq = dTC/dq. The second order (sufficient)
condition for profit maximization is d2 π/dq2 <0.That is,
the marginal profit must be decreasing at the optimal
level of q. For q less than q*, profit must be increasing,
and for q < q*, profit must be decreasing.
 Figure 13.1, Nicholson, PGE 13.1

 EXAMPLE: Nicholson Example 13.1, page 376.
235
 The fundamental condition of profit maximization (MR(q) = MC(q)) has several concrete
interpretations.
◦ For profit to be maximized, the firm must choose output level that equates MR and MC. MR = MC.
◦ The firm should hire an amount of input such that the marginal revenue from employing one more unit
of it should equal to the marginal cost of hiring that additional input.
◦ Firm of identical cost and revenue functions, in the long run the firms should have equal profits, since
each firm can imitate the actions of the other. This condition is very simple, but it is surprisingly
powerful.
 Revenue is composed of two parts: sales and prices of outputs. Costs are composed of two
parts: amount of inputs and prices of inputs. Hence, the firm’s decision then relates to what
prices it wishes to charge for its outputs or pay for its inputs, and what levels of outputs and
inputs it wishes to use.
 Hence, a firm faces two kinds of constraints:

◦ Technological constraints: constraints that concern the feasibility of the production plan
◦ Market constraints: constraints that concern the effect of actions of other agents.
 A price taking firm takes prices as given. Such a firm will be concerned only with determining
the profit maximizing levels of outputs and inputs.
236
 The profit maximization problem of the firm is π(p) = max py, such that y
is in Y.
 The function π(p) , which gives maximum profit as a function of the prices,
is called the profit function of the firm. (more on this in the next
chapter)
 The short run profit function (also known as the restricted profit
function) can be defined as π(p, z) = max py, such that y is in Y(Z).
 If the firm produces only one output, the profit function can be written as
π (p, w) = max pf(x) – wx (p is the scalar price of output, and w is the
price of inputs).
 Profit maximization of a price taking firm also implies p (∂ f(x*)/ ∂xi) = wi.
This condition says that the value of the marginal product of each
factor must be equal to its price.
237
Marginal revenue and elasticity
 Marginal revenue = MR(q) = dTR/dq = d[p(q)*q]/dq = p +
q*(dp/dq). Note that the marginal revenue is a function of
output. In general, MR will be different for different levels of q.
If price does not change as q changes,
MR = P, in which case the firm is a price taker. If price falls a q
increases (dp/dq <0) then the marginal revenue will be less
than price
 The concept of marginal revenue is directly related to the

concept of price elasticity of demand. eq,p = (dQ/Q)/ (dp/p) =
dQ/dp * p/Q. If we denote the eq,p to denote the price
elasticity of demand curve facing a single firm, we have,
MR = p + qdp/dq = p(1 + q/p * dp/dq) = p(1 + 1/ eq,p). Hence,
238
Marginal revenue and elasticity (2)
 If eq,p < -1  MR >0, i.e. if demand is elastic, MR will
be positive. If demand is elastic, the sale of one more
unit will not affect price much and hence more
revenue will be yielded by the sale.If demand facing
the firm is infinitely elastic, MR = P.
 If eq,p = -1  MR = 0, i.e. there will be no change in
revenue.
 If eq,p > -1  MR <0, i.e. if demand is inelastic,
marginal revenue will be negative Increases in sales
will be accompanied by reduction in total revenue.
239
Marginal revenue and elasticity (3)
 Since MR = MC for profit maximizing firms, we have MC = p(1 + 1/eq,p or
(p – MC)/p = -1/ eq,p. The gap between price and marginal cost will
decrease as the demand curve facing the firm becomes more elastic. For a
price taking firm (where eq,p = ∞, p = MR=MC and there is no gap.
 Since the gap between marginal cost and price is an important measure of
inefficient resource allocation, this equation is widely used in empirical
studies of market organization.
 This equation makes sense only if eq,p < -1. If eq,p were > -1, the equation
would imply a negative marginal cost- an impossibility. Hence profit
maximizing firms must operate at points on the demand curves they face
where demand is elastic.
240
Marginal revenue curve
 Any demand curve has a marginal revenue curve
associated with it, i.e marginal revenue curve cannot
be calculated without referring to specific demand
curve.
 In the usual case of a down ward sloping demand

curve, the marginal revenue curve will lie below the
revenue or demand curve since MR < p.
 (Figure 13.2, Nicholson, page 379)
241
Firm Supply
Short-run supply curve:
 The firm’s short-run supply curve shows how much it will produce
at various output prices. For a profit maximizing firm that takes the
price of output as given, the curve consists of the positively sloped
segment of the firm’s short-run marginal cost above the point of
minimum average variable cost. For prices below this level, the
firm’s profit maximization decision is to shut down and produce no
output.
 Figure 13.3 (Nicholson, page 380).
 Any factor that shifts the firm’s short-run marginal cost curve (eg.
input prices or changes in the level of fixed inputs used) will also
shift the short-run supply curve.
242
Input demand and the supply function:
 Given q = (f(K, L), the firm’s economic profits can be expressed as function of the inputs it
employs: π (K, L) = pq – TC(q) = p f(K, L) – vK – wL, where v and w are prices of capital and
labor, respectively. The first order condition for profit maximization then is: ∂π/ ∂K = p ∂f/ ∂K
– v =0 and
∂π/ ∂L = p ∂f/ ∂L – w =0 (solving this FOC, we can get the input demand
functions)
 To see the connection between supply behavior from input decision making
perspective and output decisions, we use the FOC listed above. If the optimal levels
of the inputs would be k* and L*, then we have the input demand functions as K* =
f(p, v, w), and L* = f (p, v, w). Substituting these input choices into the production
function, we get the profit maximization output choices (supply), q* = f (K*, L*) = f
(K*(p, v, w), L*(p, v, w))= q*(p, v, w).
 Supply function: The supply function for a profit maximizing firm that takes both
output prices and input prices (v, w) as given is written as q supplied = q*(p, v, w).
This shows that output choices depend on both product prices and input cost
considerations.
 The supply function provides a convenient reminder of two points that do not come
clearly from the marginal cost curve approach to supply:
(1)The firm’s output decision is fundamentally a decision about hiring inputs
(2) Changes in input costs will alter the hiring of inputs and affect output choices as well.
Example: Nicholson, Example 13.3, page 386.
243
Properties of input demand and output supply
functions
 The functions that give optimal choices of inputs and outputs as a function of the
prices are known as factor demand and output supply functions.
 The fact that these functions are the solutions to the a maximization problem of a
specific form, the profit maximization problem, implies certain restrictions on the
behaviour of demand and supply functions.
 The list of restrictions could be used to examine theoretical statements about

how a profit-maximizing firm would respond to changes in its economic
environment or to empirically test whether a particular firm’s observed behaviour
is consistent with the profit maximization model.
 Determination of the properties of input demand and output supply functions can
be done in three ways:
◦ By examining the FOC that characterize the optimal choices
◦ To examine the maximizing properties of the input demand and supply functions directly
◦ By examining the properties of the profit and cost functions and relate these properties
to the input demand functions. This is the dual approach.
244
Properties of input demand and output
supply functions (2)
Two Properties of factor demand functions:
 The factor demand functions, xi(p, w), are homogeneous
of degree zero. That is, xi(tp, tw) = xi(p, w). This
property is an important implication of the profit-
maximizing behaviour. An immediate way to check
whether some observed behaviour could come from
profit-maximizing model is to see if the input demand
functions are homogenous of degree zero. If they are
not the firm could not possibly maximizing profits.
 The factor demand curve slopes downward (with
respect to its price)
245
Application, problems and extensions
 Applications: Nicholson, pages 390-391
 Problems, Nicholson, pages 391392,
problems 13.1, 13.3, 13.4, 13.5, 13.7, 13.8
246
Chapter 4
UNCERTAINTY AND
RISK AVERSION
247
Probability
 The probability of a repetitive event

happening is the relative frequency with
which it will occur
◦ probability of obtaining a head on the fair-flip of a
coin is 0.5
 If a lottery offers n distinct prizes and the
probabilities of winning the prizes are i
(i=1,n) then
n

i 1
i 1
248
Expected Value
 For a lottery (X) with prizes x1,x2,…,xn and

the probabilities of winning 1,2,…n, the
expected value of the lottery is
E( X )  1x1  2 x2  ...  n xn
n
E ( X )   i xi
i 1
 The expected value is a weighted sum of the

outcomes
 the weights are the respective probabilities
249
Expected Value
 Suppose that Smith and Jones decide to flip a

coin
◦ heads (x1)  Jones will pay Smith $1
◦ tails (x2)  Smith will pay Jones $1
 From Smith’s point of view,
E( X )  1x1  2 x2
1 1
E ( X )  ($1)  ( $1)  0
2 2
250
Expected Value
 Games which have an expected value of zero

(or cost their expected values) are called
actuarially fair games
◦ a common observation is that people often refuse
to participate in actuarially fair games
251
Fair Games
 People are generally unwilling to play fair

games
 There may be a few exceptions
◦ when very small amounts of money are at stake
◦ when there is utility derived from the actual
play of the game
 we will assume that this is not the case
252
St. Petersburg Paradox
 A coin is flipped until a head appears

 If a head appears on the nth flip, the player
is paid $2n
x1 = $2, x2 = $4, x3 = $8,…,xn = $2n
 The probability of getting a head on the ith
trial is (½)i
1=½, 2= ¼,…, n= 1/2n
253
St. Petersburg Paradox
 The expected value of the St. Petersburg

paradox game is infinite
  i
 1
E ( X )   i xi   2  
i
i 1 i 1 2
E ( X )  1  1  1  ...  1  
 Because no player would pay a lot to play
this game, it is not worth its infinite expected
value
254
Expected Utility
 Individuals do not care directly about the dollar

values of the prizes
◦ they care about the utility that the dollars provide
 If we assume diminishing marginal utility of
wealth, the St. Petersburg game may converge
to a finite expected utility value
◦ this would measure how much the game is worth to
the individual
255
Expected Utility
 Expected utility can be calculated in the same

manner as expected value
n
E ( X )   i U ( xi )
i 1
 Because utility may rise less rapidly than

the dollar value of the prizes, it is possible
that expected utility will be less than the
monetary expected value
256
The von Neumann-Morgenstern
Theorem
 Suppose that there are n possible prizes
that an individual might win (x1,…xn)
arranged in ascending order of desirability
◦ x1 = least preferred prize  U(x1) = 0
◦ xn = most preferred prize  U(xn) = 1
257
Theorem
 The point of the von Neumann-Morgenstern
theorem is to show that there is a
reasonable way to assign specific utility
numbers to the other prizes available
258
Theorem
 The von Neumann-Morgenstern method is
to define the utility of xi as the expected
utility of the gamble that the individual
considers equally desirable to xi
U(xi) = i · U(xn) + (1 - i) · U(x1)
259
Theorem
 Since U(xn) = 1 and U(x1) = 0
U(xi) = i · 1 + (1 - i) · 0 = i
 The utility number attached to any other
prize is simply the probability of winning it
 Note that this choice of utility numbers is
arbitrary
260
Expected Utility Maximization
 A rational individual will choose among

gambles based on their expected utilities
(the expected values of the von Neumann-
Morgenstern utility index)
261
 Consider two gambles:
◦ first gamble offers x2 with probability q and x3
with probability (1-q)
expected utility (1) = q · U(x2) + (1-q) · U(x3)
◦ second gamble offers x5 with probability t and x6
with probability (1-t)
expected utility (2) = t · U(x5) + (1-t) · U(x6)
262
 Substituting the utility index numbers
gives
expected utility (1) = q · 2 + (1-q) · 3
expected utility (2) = t · 5 + (1-t) · 6
 The
individual will prefer gamble 1 to
gamble 2 if and only if
q · 2 + (1-q) · 3 > t · 5 + (1-t) · 6
263
 If individuals obey the von Neumann-
Morgenstern axioms of behavior in
uncertain situations, they will act as if they
choose the option that maximizes the
expected value of their von Neumann-
Morgenstern utility index
264
Risk Aversion
 Two lotteries may have the same expected

value but differ in their riskiness
◦ flip a coin for $1 versus $1,000
 Risk refers to the variability of the outcomes
of some uncertain activity
 When faced with two gambles with the same
expected value, individuals will usually choose
the one with lower risk
265
Risk Aversion
 In general, we assume that the marginal utility

of wealth falls as wealth gets larger
◦ a flip of a coin for $1,000 promises a small gain in
utility if you win, but a large loss in utility if you
lose
◦ a flip of a coin for $1 is inconsequential as the gain
in utility from a win is not much different as the
drop in utility from a loss
266
Risk Aversion
 Suppose that the person is offered two
fair gambles:
◦ a 50-50 chance of winning or losing $h
Uh(W*) = ½ U(W* + h) + ½ U(W* - h)
◦ a 50-50 chance of winning or losing $2h
U2h(W*) = ½ U(W* + 2h) + ½ U(W* - 2h)
267
Risk Aversion
 The person will prefer current wealth to

that wealth combined with a fair gamble
 The person will also prefer a small gamble
over a large one
268
Risk Aversion and Insurance
 The person might be willing to pay some

amount to avoid participating in a gamble
 This helps to explain why some individuals
purchase insurance
269
Risk Aversion and Insurance
 An individual who always refuses fair bets
is said to be risk averse
◦ will exhibit diminishing marginal utility of
income
◦ will be willing to pay to avoid taking fair bets
270
Willingness to Pay for Insurance
 Consider a person with a current wealth of
$100,000 who faces a 25% chance of losing
his automobile worth $20,000
 Suppose also that the person’s von
Neumann-Morgenstern utility index is
U(W) = ln (W)
271
 The person’s expected utility will be

E(U) = 0.75U(100,000) + 0.25U(80,000)
E(U) = 0.75 ln(100,000) + 0.25 ln(80,000)
E(U) = 11.45714
 In this situation, a fair insurance premium

would be $5,000 (25% of $20,000)
272
 The individual will likely be willing to pay more

than $5,000 to avoid the gamble. How much will
he pay?
E(U) = U(100,000 - x) = ln(100,000 - x) = 11.45714
100,000 - x = e11.45714
x = 5,426
 The maximum premium is $5,426
273
Measuring Risk Aversion
 The most commonly used risk aversion

measure was developed by Pratt
U " (W )
r (W )  
U ' (W )
 For risk averse individuals, U”(W) < 0
 r(W) will be positive for risk averse individuals
 r(W) is not affected by which von Neumann-
Morganstern ordering is used
274
 The Pratt measure of risk aversion is

proportional to the amount an individual
will pay to avoid a fair gamble
275
 Let h be the winnings from a fair bet

E(h) = 0
 Let p be the size of the insurance premium
that would make the individual exactly
indifferent between taking the fair bet h and
paying p with certainty to avoid the gamble
E[U(W + h)] = U(W - p)
276
 We now need to expand both sides of the

equation using Taylor’s series
 Because p is a fixed amount, we can use a
simple linear approximation to the right-
hand side
U(W - p) = U(W) - pU’(W) + higher order terms
277
 For the left-hand side, we need to use a

quadratic approximation to allow for the
variability of the gamble (h)
E[U(W + h)] = E[U(W) - hU’(W) + h2/2 U”(W)

+ higher order terms
E[U(W + h)] = U(W) - E(h)U’(W) + E(h2)/2 U”(W)
+ higher order terms
278
 Remembering that E(h)=0, dropping the

higher order terms, and substituting k for
E(h2)/2, we get
U (W )  pU ' (W )  U (W )  kU " (W )
kU " (W )
p  kr (W )
U ' (W )
279
Risk Aversion and Wealth
 It is not necessarily true that risk aversion

declines as wealth increases
◦ diminishing marginal utility would make potential
losses less serious for high-wealth individuals
◦ however, diminishing marginal utility also makes the
gains from winning gambles less attractive
 the net result depends on the shape of the utility function
280
 If utility is quadratic in wealth

U(W) = a + bW + cW 2
where b > 0 and c < 0
 Pratt’s risk aversion measure is
U " (W )  2c
r (W )   
U ' (W ) b  2cW
 Risk aversion increases as wealth
increases
281
 If utility is logarithmic in wealth

U(W) = ln (W )
where W > 0
U " (W ) 1
r (W )   
U ' (W ) W
 Risk aversion decreases as wealth
increases
282
 If utility is exponential
U(W) = -e-AW = -exp (-AW)
where A is a positive constant
U " (W ) A2e  AW
r (W )     AW
A
U (W ) Ae
 Risk aversion is constant as wealth
increases
283
Relative Risk Aversion
 It seems unlikely that the willingness to
pay to avoid a gamble is independent of
wealth
 A more appealing assumption may be that
the willingness to pay is inversely
proportional to wealth
284
 This relative risk aversion formula is
U " (W )
rr (W )  Wr (W )   W
U ' (W )
285
 The power utility function

U(W) = WR/R for R < 1,  0
exhibits diminishing absolute relative risk
aversion
U " (W ) (R  1)W R 2 (R  1)
r (W )    R 1

U ' (W ) W W
but constant relative risk aversion
rr (W )  Wr (W )  (R  1)  1  R
286
The State-Preference Approach
 The approach taken in this chapter up to
this point is different from the approach
taken in other chapters
◦ has not used the basic model of utility-
maximization subject to a budget constraint
 There is a need to develop new techniques
to incorporate the standard choice-
theoretic framework
287
States of the World
 Outcomes of any random event can be

categorized into a number of states of the
world
◦ “good times” or “bad times”
 Contingent commodities are goods
delivered only if a particular state of the
world occurs
◦ “$1 in good times” or “$1 in bad times”
288
States of the World
 It is conceivable that an individual could

purchase a contingent commodity
◦ buy a promise that someone will pay you $1 if
tomorrow turns out to be good times
◦ this good will probably cost less than $1
289
Utility Analysis
 Assume that there are two contingent goods

◦ wealth in good times (Wg) and wealth in bad times
(Wb)
◦ individual believes the probability that good times
will occur is 
290
Utility Analysis
 The expected utility associated with these

two contingent goods is
V(Wg,Wb) = U(Wg) + (1 - )U(Wb)
 This is the value that the individual wants to
maximize given his initial wealth (W)
291
Prices of Contingent Commodities
 Assume that the person can buy $1 of wealth
in good times for pg and $1 of wealth in bad
times for pb
 His budget constraint is
W = pgWg + pbWb
 The price ratio pg /pb shows how this person
can trade dollars of wealth in good times for
dollars in bad times
292
Fair Markets for Contingent Goods
 If markets for contingent wealth claims are
well-developed and there is general
agreement about , prices for these goods
will be actuarially fair
pg =  and pb = (1- )
 The price ratio will reflect the odds in favor
of good times
pg 

pb 1  
293
Risk Aversion
 If contingent claims markets are fair, a

utility-maximizing individual will opt for a
situation in which Wg = Wb
◦ he will arrange matters so that the wealth
obtained is the same no matter what state
occurs
294
Risk Aversion
 Maximization of utility subject to a budget

constraint requires that
V / Wg U ' (Wg ) pg
MRS   
V / Wb (1  )U ' (Wb ) pb
 If markets for contingent claims are fair
U ' (Wg )
1
U ' (Wb )
Wg  Wb
295
Insurance in the State-Preference
Model
 Again, consider a person with wealth of
$100,000 who faces a 25% chance of losing his
automobile worth $20,000
◦ wealth with no theft (Wg) = $100,000 and
probability of no theft = 0.75
◦ wealth with a theft (Wb) = $80,000 and probability
of a theft = 0.25
296
Model
 If we assume logarithmic utility, then
E(U) = 0.75U(Wg) + 0.25U(Wb)
E(U) = 0.75 ln Wg + 0.25 ln Wb
E(U) = 0.75 ln (100,000) + 0.25 ln (80,000)
E(U) = 11.45714
297
Model
 The budget constraint is written in terms of
the prices of the contingent commodities
pgWg* + pbWb* = pgWg + pbWb
 Assuming that these prices equal the
probabilities of these two states
0.75(100,000) + 0.25(80,000) = 95,000
 The expected value of wealth = $95,000
298
Model
 The individual will move to the certainty line
and receive an expected utility of
E(U) = ln 95,000 = 11.46163
◦ to be able to do so, the individual must be able to
transfer $5,000 in extra wealth in good times into
$15,000 of extra wealth in bad times
 a fair insurance contract will allow this
 the wealth changes promised by insurance (dWb/dWg) =
15,000/-5,000 = -3
299
A Policy with a Deductible
 Suppose that the insurance policy costs

$4,900, but requires the person to incur the
first $1,000 of the loss
Wg = 100,000 - 4,900 = 95,100
Wb = 80,000 - 4,900 + 19,000 = 94,100
E(U) = 0.75 ln 95,100 + 0.25 ln 94,100
E(U) = 11.46004
 The policy still provides higher utility than
doing nothing
300
Risk Aversion and Risk Premiums
 Consider two people, each of whom starts
with an initial wealth of W*
 Each seeks to maximize an expected utility
function of the form
WgR WbR
V (Wg ,Wb )    (1  )
R R
 This utility function exhibits constant
relative risk aversion
301
Risk Aversion and Risk Premiums
WgR WbR
V (Wg ,Wb )    (1  )
R R
 The parameter R determines both the degree

of risk aversion and the degree of curvature
of indifference curves implied by the function
◦ a very risk averse individual will have a large
negative value for R
302
 The formal explanation of decision-making under risk and uncertainty
began with the development of the expected utility theory (Von-Neuman
and Morgestern, 1944).
 This theory is based on certain axioms of individual’s behaviour in making
choices among risky prospects and hypothesizes the existence of a utility
function that assigns cardinal values to random outcomes based on the
preference of the decision-maker.
 Expected utility theory assumes that the preferences of the decision
maker comply with the axioms of ordering and transitivity, continuity and
independence, and that there is a utility function U that assigns a numerical
value to each alternative.
 In doing so, it allows the ranking of alternatives within the risk context.
 The expected utility value of an uncertain prospect weighted by its
probability of occurrence gives the expected utility index of the decision-
maker.
 The objective function of the decision maker is to rank prospects
according to their probability of occurrence and thus the expected utility
index.
303
•Compliance with the behavioural axioms of the utility theory
does not restrict an individual's utility function to any
particular functional form.
•However, the risk preference (attitude) of an individual is
determined by the functional form of the utility function.
Based on the curvature of the utility function, individuals can
be classified as risk averse (concave utility function), risk
loving (convex utility function), and risk neutral (linear utility
function). A risk averse (loving) person is willing to give up
part of his wealth (income) to avoid (take) risk, respectively.
The amount a risk averse individual is willing to pay to avoid
risky outcomes is called the risk premium (R). Given any
prospect X, the certainty equivalent (CE) is the amount an
individual is willing to accept as insurance against a risky
alternative to the sure outcome,
(1)
304
Where V is the standard deviation of the
prospect X, and ra is the absolute risk averse
coefficient. Based on equation 1, the risk
premium is greater, less than or equal to zero for
risk averse, risk loving and risk neutral
individuals, respectively.
(2)
<CE
= CE
According to equation 2, the utility of expected outcome is
greater than, less than or equal to the certainty equivalent for
risk averse, risk lover and risk neutral individuals, respectively. 305
 Based on the utility function, several measures
of risk aversion can be observed. For example
if U`>0, more is better than less. If U``>0, the
decision maker is risk lover. If U``<0, the
decision maker is risk averse and finally if
U``=0, the decision maker is risk neutral.
 The relationship between the shape of the
utility function, risk attitude, risk premium and
marginal utility is presented in table 1.
306
Table 1 Relationships between the shape of the utility function, risk
attitude, risk premium and marginal utility
Shape of the Attribute

utility
function U(X)
Risk attitude Risk premium Marginal utility
(EV - CE) (dU/dX>0)
Concave aversion positive diminishing
(EV > CE) (d2U/dX2 < 0)
Linear neutrality zero constant
(EV = CE) (d2U/dX2 = 0)
Convex preference negative increasing
(EV < CE) (d2U/dX2 > 0)
307
Various alternative measures of risk aversion have been developed to
describe risk preference of economic agents. Pratt (1964) and Arrow
(1965), developed absolute risk aversion (ra) which is the most
commonly used measures of risk preference of individuals. The
formula of absolute risk aversion is given by:
(3)
This coefficient can be interpreted as the percentage change in marginal utility
caused by each monetary unit of gain or loss (Raskin and Cochram, 1986). Thus,
the coefficient ra takes either positive, negative or zero values for risk-loving, risk-
averse or risk neutral economic agents, respectively. When the coefficient decreases
as the monetary value increases we have decreasing absolute risk aversion (DARA).
Alternatively, if the coefficient increases under the same set of circumstances we
have increasing absolute risk aversion (IARA). Finally, if the coefficient does not
change across the monetary level, the decision maker exhibits constant absolute risk
aversion (CARA), which implies that the level of the argument of the utility
function does not affect his or her decisions under uncertainty.
308
Since ra is not a non-dimensional measure of risk aversion,
its value is dependent on the currency in which the monetary
units are expressed. To overcome the impossibility of
comparing risk aversion among different economic agents
Arrow (1965) and Pratt (1964) have devised a
nondimensional measure; the relative risk aversion
coefficient (rr) which is given by:
(4)
This coefficient measures the percentage change of marginal utility in terms of the
percentage change in the monetary variable; hence, rr represent the elasticity of
the marginal utility function, which ranges from slightly risk-averse to extremely
risk-averse. As with the absolute risk aversion coefficient, we can find decreasing,
constant or increasing relative risk-aversion behaviour (DRRA, CRRA and IRRA,
respectively).
309
 In uncertain situations, individuals are

concerned with the expected utility
associated with various outcomes
◦ if they obey the von Neumann-Morgenstern
axioms, they will make choices in a way that
maximizes expected utility
310
 If we assume that individuals exhibit a

diminishing marginal utility of wealth, they
will also be risk averse
◦ they will refuse to take bets that are actuarially
fair
311
 Risk averse individuals will wish to insure

themselves completely against uncertain
events if insurance premiums are
actuarially fair
◦ they may be willing to pay actuarially unfair
premiums to avoid taking risks
312
 Decisions under uncertainty can be analyzed

in a choice-theoretic framework by using
the state-preference approach among
contingent commodities
◦ if preferences are state independent and prices
are actuarially fair, individuals will prefer
allocations along the “certainty line”
 will receive the same level of wealth regardless of
which state occurs
313
Chapter 5
Transaction cost economics and
property right theory
1/30/2022 Hassen Beshir (PhD) 314

Chapter 5: Theories of Institutional
Change
 Competing theories
 Basic questions:
a) What are the reasons for institutional change?
b) Who are the key actors involved in the processes of institutional

change?
c) Is institutional change induced or imposed?
Hassen Beshir (PhD)

1/30/2022 315
Components of the NIE
Public Choice & Political Economy

(Buchanan, Tullock, Olson, Bates) Social Capital
(Putnam, Coleman)
NIE
Property rights literature
(Alchian, Demsetz)
Transaction Costs Economics
(Coase, North, Williamson)
Economics of information
(Akerlof, Stigler, Stiglitz)
Theory of Collective Action
(Ostrom, Olson, Hardin)
Law and Economics

(Posner)
5.1. Theory of Induced Institutional
Innovation
 The driving forces for institutional change is the
technical progress and change in relative factor
prices (Hayami and Ruttan 1985)
 Institutional change is the product of the

interaction between demand and supply,
basically influenced by the neoclassical
equilibrium model. Market forces create a
demand for institutional change.

5.2 Transaction cost economics
 According to this theory, institutional change takes place
to reduce transaction costs, including:
◦ Cost of screening and selecting a buyer or seller
◦ Cost of obtaining information on the good or service
◦ Cost of bargaining & negotiating a contract
◦ Cost of monitoring & enforcing the contract

 Williamson (1996, 2000):
◦ Combines the concepts of bounded rationality & opportunistic

behavior to explain contracts & ownership structure of firms
◦ Continuum of organizational form (from vertical integration to cash

markets) that depends largely on the magnitude of transaction
costs

 North (1986, 1989, 1994)
◦ Institutions that evolve to reduce transaction costs are key to the
performance of economies
◦ Not all institutions that emerge are efficient
◦ Role of government is crucial in specifying property rights and enforcing
contracts
 North (1990)
“The inability of societies to develop effective, low-cost enforcement of
contracts is the most important source of both historical stagnation and
contemporary underdevelopment in the third world.”

How is transaction cost economics relevant?
Globalization & Market liberalization &

industrialization of government
world agriculture devolution
Increasing reliance on vertical linkages, long-

term contracts, and coordinated relationships

Characteristics of rural agricultural- economy in developing countries:
 Small farmers and traders face high transaction costs resulting in thin
markets
 Market failure in the provision of credit, inputs, and services in remote

areas
 Incomplete or imperfect land and labor markets

 The transaction costs literature will be important in:
◦ Explaining the choice of contracts between different market

participants
◦ Analyzing the type of institutional innovation needed to integrate small

farmers and the poor in the new agricultural economy
◦ Understanding the role of the government and the private sector in

supporting the development of these institutions

Example: Contract farming
 Contract farming as a way to cut transaction costs and
include small farmers in high-value markets
 Questions:
◦ What are the conditions that make contract farming

sustainable and beneficial to small and poor farmers?
◦ What is the role of the government in improving those

conditions?

5.3 Distributive bargaining theory
 The driving forces for institutional change is distributional

conflict over resources of significant economic importance
(Libecap, 1989: Knight 1992, 1997).
 In this theory, institutions are the by-products of social

conflict
 Institutional change is a response to conflict to correct for

distributional imbalances
 Take an example of property rights change

5.3 Distributive bargaining
theory…
 To apply bargaining theory, one needs to allow for the possibility
that some social actors are more powerful than others and
investigates the effect of those differences (Knight 1992:127).
 The success of an actor in bargaining is directly related to the

ability of an actor to produce strategic commitments (or threats),
i.e. compliance or non-compliance to the rules of the game.
 The theory can be applied where there is conflict over resource

uses between groups or individuals.

5.3 Distributive bargaining theory
if P,AP < x, there will be two equilibrium outcomes.
If P chooses strategy G and AP chooses C, then P gains more than AP.

This difference (P > 0) is known as ‘distributional advantage’ and each
group tries to achieve .

5.4. Political-Economy Theory
5.4.1 Interest Groups
 there is a political exchange between interest groups (Olson 1982)
 “Politicians hear nothing from many, but a lot from few”;
 Dedicated group of voters influence processes of institutional

change in return for their political support
 rent-seekers emerge in the process, having access to public fund in

a very unnoticeable way to the public

5.4. Political-Economy Theory
5.4.2 Public choice theory
 James Buchanan came up with public choice theory
 the principal-agent relationship between the voters and

politicians (Tullock 1987).
 Thus, institutional change arises from the influence of

the principals (constituents) on their agents who are
expected to maintain their promises or commitments
(political market)

5.5 Property rights theory
5.5.1 Concept of Property Rights
 many people regard property as a tangible ‘physical object’. Institutional

economists use a different conceptual language.
 Property is considered as a “benefit (or income) stream” in that the owner

controls this benefit stream (Bromley 1991).
 A right may be a ‘set of actions and behaviors that possessor of a property

may not be prevented from undertaking, or a duty on all others to refrain
from preventing those actions or behaviors’.
 Therefore, rights are not relationships between the right holder and an
object, but rather are relations between the right holder and other people
with respect to the object (Bromley 1991:15).

5.5 Property rights theory…
Alternative definitions
1. property right is defined as “the capacity to call upon

the collective to stand behind one’s claim to benefit
stream” from an asset of economic importance
(Bromley 1991:15).
2. “ property rights over assets consists of the rights, or

the powers, to consume, obtain income from, and
alienate these assets” Barzel (1989:2).

5.5 Property rights theory…
3. “bundles of rights” including access and withdrawal,
exclusion, management and alienation rights (Schlager
and Ostrom 1992: 249-251)
4. Furubotn and Richter (2005) review and classify

property rights as “absolute” and “relative”, the
former involving assignment of exclusive individual
property rights to physical objects, whereas “relative”
property rights include all rights related to contractual
obligations and agreements
Relative PRs: Exchange. Contractual obligations -

particularly when time is an important part of the
“transaction” -> monitoring and enforcement of the
contract
5.5.2 Property rights regimes
1) Private property:
- is any property that is not public property. Private property may be under the
control of a single individual or by a group of individuals collectively.
- a set of ordered institutional arrangements in which the state protects the

rights of certain individuals to access, control and manipulate resource benefit
steams. Others have a right to expect that only socially-acceptable uses will
occur, and a duty to refrain from preventing those uses.
- What are the advantages of private property? Some agree on its role in the
efficient allocation of resources and investment in resource improvements
- Why have claims to private land use emerged in pastoral and agropastoral
areas? (e.g. In the case of priavte ranches among the Maasai)

5.5.2 Property rights regimes…
2) Common Property: distinction of “common-pool” and “common property”.
 a common-pool resource (CPR) is a particular type of good consisting of a natural

or human-made resource system, the size or characteristics of which makes it
costly, but not impossible, to exclude potential beneficiaries from obtaining benefits
from its use.
 Common Property is the private property of a group of co-owners.
 "common property regime" refers to a particular social arrangement regulating the

preservation, maintenance, and consumption of a common-pool resource.
 common-pool resources are not necessarily governed by common property

regimes, they may be owned by national, regional or local governments as public
goods, by communal groups as Common property resources, or by private
individuals or corporations as private goods.

 Property rights in productive resources – land itself,
trees, pasture, many water sources, fish, etc. often
characterized by non-private property rights structures
 Substantial part of income generation, especially in SSA
 Poor, and reliance on non-private resources as safety net
 Almost all “big” environmental problems are a function

of resources under non-private, non-market property
rights (desertification, forest management, soil erosion,
pollution, over-fishing, overgrazing)

2) Common property…
 Examples of common-pool resources include irrigation systems, fishing

grounds, pastures, and forests
 common-pool resources are generally subject to the problems of

congestion, overuse, pollution, and potential destruction unless harvesting
or use limits are devised and enforced, which could lead to the
deterioraton of livelihoods.
 in a common property regime, a common-pool resource has the

appearance of a private good from the outside and that of a common
good from the point of view of an insider. The resource units withdrawn
from the system are typically owned individually by the appropriators. A
common property good is rivaled in consumption.

Design principles for a long-enduring CPR institutions
1. Clearly defined boundaries

2. Congruence between appropriation and provision rules
3. Collective-choice arrangements allowing for the participation of most of the
appropriators in the decision making process
4. Effective monitoring by monitors who are part of or accountable to the
appropriators

Design principles for a long-enduring CPR institutions
5. Graduated sanctions for appropriators who do not respect
community rules
6. Conflict-resolution mechanisms which are cheap and easy
7. Minimal recognition of rights to organize (e.g., by the
government)
8. In case of larger CPRs, Organisation in the form of multiple
layers of nested enterprises

3) State Property:
- Resources are nationalized and citizens may have use rights, while
the state has all forms of rights to the resources in question.
- a state property regime is a set of institutional arrangements in

which the state retains direct control of the benefits derived from a
resource by determining access and use rules for individuals
(Bromely, 1989, 1991).
- In many socialist countries, individuals are entitled to use resources

but not to transfer rights to the resources without the interference
of the state.

However, such state property regimes fail due to:
 rigidity of the state agencies in their application of rules;
 state agencies usually ignore, or even attempt to undermine, indigenous

political structures and institutions;
 state agencies often lack the power, authority and/or will to implement
rules prescribed at regional or national levels; and,
 state employees who are responsible for the enforcement of resources

use rules are often remunerated, legally or illegally, through the collection
of fines. (e.g. corruption and bribes)

4) open access:
 When resources are owned by no one or are used by
all without any restriction, they exihibit open access
resources
 Such lack of property regime leads to resource

destruction and increases behavioral uncertainty
 Often termed as “tragedy of open access” or formerly

coined as “the tragedy of the commons” (G. Hardin
1968)

.5.2 Property rights regimes…
Bundles of rights
 Note: we do not find very much distinction in property

rights as such, rather a bundle of rights approach is
important.
 The bundle of rights is a common way to explain the

complexities and interdependence of rights to
resources among actors having different positions


5.5.3 Forces of change in property rights
1) Equity versus efficiency
 There is tradeoff in achieving economic efficiency and attaining equity

(distribution); one is satisfied at the expense of the other, i.e. efficiency
cannot be materialized when distributional inequality is at stake (Wang
2001; Eggertsson 1990).
 The distribution view recognizes that the forces of change in property

rights are the inherent dissatisfactions by certain groups or individuals
from the existing property rights structure. The existing distribution of
property rights may benefit some but harms others.
 In such contexts, change is determined by the capacity of actors to

‘contract’ for property rights change or persistence of existing ones
(Libecap 1989).

5.5.3 Forces of change in property rights
 the economic efficiency - property rights change results from opening of
new markets, change in relative prices of factors of production,
demographic shift and technological innovation.
 change in these factors in an economy creates a pressure for change in

property rights (Demsetz 1967; Boserup 1965; North 1981; Bromley 1991;
Ensminger 1997).
 They argue that efficiency of land use increases when property rights
change from a purely open access (no property rights) to common and
then to private due to population growth and resource scarcity leading to
more commercialization.
 Though transaction costs of enforcing rights are increasing at each stage,

the return from improved efficiency largely outweighs the costs (Bromley
1991)

5.5.3 Forces of change in property rights…
2) The ‘ cost-benefit’ versus ‘scarcity-incursion’ analysis
a) Cost-benefit argument –
• property rights emerge when the benefits obtained from controlling access
to resources exceed the transaction costs of defending the resource from
others, i.e. the ‘social cost-benefit’ comparisons or ‘internalization of
externalities’ from introducing new property rights.
• gains from internalization of externalities must exceed the costs to cause

alteration of property rights regime. Those economic forces fetching new
opportunities (e.g. emergence of markets or newly introduced technology)
should ensure cost-effective way of internalizing external effects (Demsetz
1967) (e.g. controlling of hunting areas as a result of the rise of fur
industry)

b) the scarcity-incursion analysis - a counterargument challenging
Demsetz’s view and that of his proponents (Field 1989). This is
particularly the case in terms of predicting the direction of change in
property rights when the value of a resource rises.
An increase in resource value would rather lead to less exclusive

property rights since the more valuable resource will attract greater
incursion (Field 1989).
According to Field, a higher resource value will cause pressure from

outsiders intending to use the resource. This will cause the
exclusion cost to rise much more than the benefits obtained from
excluding others.


 An increase in the number of these commons (from a single large commons to

completely divided parcels for individuals N) will lead to a decline in the marginal
governance cost (Tm) because of reduced cost in organizing collective action.
 However, such an increase in the number of commons will lead to a rise in the
marginal cost of exclusion (boundary management, Em), where an increasing
exclusion costs saves transaction costs.
 A point where marginal governance costs and marginal exclusion costs intersect
(M*), one may find an optimal number of parcels for the commons and size of
users.

 An increase in resource value (EmR) and an introduction

of new technology of exclusion (EmT) will again affect the
optimality, where both having the opposite effect, but
again maintaining the optimal point.
 Field’s model does not completely deny the influence of

population and economic growth as determining factors
for private property to emerge.

5.5.4 Property rights and legal pluralism (LP)
 state (or statutory) law as made by legislatures and enforced by the government;
 religious law, including both law based on written doctrines and accepted
religious practice;
 customary law, which may be formal written custom or living interpretations of

custom;
 project (or donor) law, including regulations associated with particular projects
or programs, such as an irrigation project;
 organizational law, such as rules made by user groups; and
 a range of local norms, which may incorporate elements of other laws.

 Thus, LP can be a possible source of conflict and uncertainty

5.6. Evolutionary theory
• Building upon an evolutionary perspective, the concept of ‘convention’
was developed based on the works of Sugden (1989), Harsanyi and
Selten (1988) and Hayek (1979).
• It refers to unintended consequences of human interaction, driven by

‘group’ other than ‘individual’ interests.
• A useful point in this theoretical argument is that though conventions

can in a stronger sense be converted to rules, they are not the result of
any collective choice and do not originate from abstract rational
analysis (Sugden 1989:97).
• It includes rules that have never been consciously designed but are in
the interest of everyone to keep (North 1990).
• Rules evolve in unintended way rather being designed on calculative basis

contrary to the rational choice.

Chapter 6
Market Structures
Dr Hassen B Market Structure and their

1/30/2022 optmization 353
Market Structures:The Degree of
Competition
 Classifying markets
– number of firms
– freedom of entry to industry
– nature of product
– nature of demand curve
 The four market structures
◦ perfect competition
◦ monopoly
◦ monopolistic competition
◦ oligopoly

SUMMARY OF MARKET STRUCTURES
Perfect Monopolistic Oligopoly Monopoly
Competition Competition
ASSUMPTIONS ABOUT:
Number of Firms Very Many Many Few One

Output of Different Firms Identical Differentiated Identical or —
Differentiated
View of Pricing Price taker Price setter Price setter Price setter
Barriers to Entry or Exit? No No Yes Yes
Strategic Interdependence? No No Yes —
PREDICTIONS:
Price and Output Decisions MC =MR MC = MR Through strategic MC = MR
Interdependence
Short-Run Profit Positive, zero, or Positive, zero, or Positive, zero, or Positive, zero, or
negative negative negative negative
Long-Run Profit Zero Zero Positive or zero Positive or zero
Advertising? Never Almost always Yes, if differentiated Sometimes
product
Example Fruit stall Corner shop Cars Post office

Profit Maximization
 Determining the profit maximizing level of output
 Profit = Π = Total Revenue -Total Cost
 Total Revenue = Pq
 Total Cost = C(q)
 Therefore:
 Profit Maximization
 As firm expands q,TR increases with units sold
 But TR increases at a decreasing rate because prices have to be cut to generate
extra sales
 TC increases at an increasing rate because of law of diminishing marginal
returns
 Expand as long as extra TR > extra TC
 Profit Maximization
 Maximize Π
 When
 dΠ/dq = 0 = q(dp/dq) + p(q) – dC/dq
 q(dp/dq) + p(q) = MR(q)
 dC/dq= MC(q) Dr Hassen B Market Structure and their
 At q*, MR(q) = MC(q)
 1
MR  P1    1
 1
MR1  P1 1  
 e  e1  MR2  P2 1  
 e2 

If e1 = e2 the ratio of prices is equal to unity.
P1
 1  P1  P2
P2
This implies that when elasticities are same, price discrimination is not possible.
The monopolist will charge uniform price for his product.
But if elasticities are different.
 1  1 
 1 
P1 1    P2 1 
1 
 1    1  
 e1   e2  →If e1 > e2, then  e1   e2 
This means P1 < P2
The market with higher elasticity will have the lower price.

dR = p.dy + ydp
dR/dy = p + dp/dy.y
Competitive firm : price equals marginal cost

THE SOURCES OF MONOPOLY
 In a perfectly competitive market, there are no significant barriers to entry
by new firms. Monopoly, by contrast, arises because of barriers to entry. In
this section, we consider the three most common barriers responsible for
creating and maintaining monopoly markets: economies of scale, control of
a scarce input, and barriers created by government.
 1) ECONOMIES OF SCALE: A natural monopoly exists when, due to
economies of scale, one firm can produce at a lower cost per unit than can two
or more firms
 2) CONTROL OF SCARCE INPUTS
 Some firms maintain their monopoly status by controlling a scarce input
needed to produce a good. For example, from 1893 until the 1940s, Alcoa
(the Aluminum Company of America) was the sole seller of aluminum in
the United States because it owned virtually all of the country’s deposits
of bauxite—a natural resource needed to produce aluminum. Similarly,
since the 1880s, De Beers, a South African company, has enjoyed a near
monopoly on the sale of finished diamonds by buying up most of the
world’s diamond mines or the raw diamonds that come from them.


 3) GOVERNMENT-ENFORCED BARRIERS
 The two main methods of creating a monopoly are (1) the protection of intellectual
property through patents, trademarks, and copyrights and (2) exclusive government
franchises.
 A) Protection of Intellectual Property..
 In dealing with intellectual property, government strikes a compromise: It allows the creators of
intellectual property to enjoy a monopoly and earn economic profit, but only for a limited period
of time. Once the time is up, other sellers are allowed to enter the market, and it is hoped that
competition among them will bring down prices
 The two most important kinds of legal protection for intellectual property are patents and
copyrights. New scientific discoveries and the products that result from them are protected
by a patent obtained from the federal government. The patent prevents anyone else from
selling the same discovery or product for about 20 years.
 Literary, musical, and artistic works are protected by a copyright, which grants exclusive
rights over the material for at least 50 years. For example, the copyright on a book is owned
by a certain Publishing. No other company or individual can print copies and sell them to the
public, and no one can quote the book at length without obtaining its permission.
 B) Government Franchise. The large firms we usually think of as monopolies -utility,
telephone, and cable television companies—have their monopoly status guaranteed through
government franchise—a grant of exclusive rights over a product. Here, the barrier to
entry is quite simple: Any other firm that enters the market will be prosecuted!
 Governments usually grant franchises when they think the market is a natural monopoly. In
this case, a single large firm—enjoying economies of scale—would
Dr Hassenhave a lower
B Market cost and
Structure pertheir
unit
than multiple smaller firms, so government tries to serve theoptmization
1/30/2022 public interest by ensuring that 363
Barriers to entry summary
– economies of scale
– product differentiation and brand
loyalty
– ownership/control of key factors
– ownership/control over outlets
– legal protection
– mergers and takeovers
– aggressive tactics and intimidation
Equilibrium of industry under perfect competition and monopoly:
with the same MC curve
P1
P2
O Q1 Q2

(Dis)advantages of monopoly
 Disadvantages of monopoly
– high prices / low output: short run
– high prices / low output: long run
– lack of incentive to innovate
 Advantages of monopoly
– economies of scale
– profits can be used for investment
– high profits encourage risk taking

• Competitive industry operates where Price
equals MC
• Monopolist can raise profit by charging
prices greater than MC
• Prices are higher and outputs lower under
monopoly
• Consumers are worse off in monopoly
• Firms are better off
• This could be examined by using the
concept dead-weight loss

Monopolist Practices
 • Price discrimination–Monopolist can
raise profit by charging different prices
in different markets
◦ –Not related to cost
◦ –Control power over price
◦ –Elasticities in separate markets differ
◦ –The monopolist appropriates more of the
consumer surplus

First degree price discrimination
The extreme situation:
– -Different price for
each unit
– Individuals
willingness to pay
– Demand curve MR
– Profit increased
• In single price case: Qm
, Pm – Total profit = AMN
• FDPD: Qc, Pc – Total
profit = AMW
– Additional profit = ANW

Second degree price discrimination
• Different prices
charged for
different blocks
(quantities)
• Block prices: P1 ,
P2 , P3
•MR  continuous
step function:
P1MLXYW
•CSAP1M+MLX+X
YW
• Profit increases by
less than ANW
Third degree price discrimination
Separate the market in to groups

• Produce in one location, sells for
separate groups
• Conditions: – Existence of barrier –
Difference in elasticity
• One marginal cost
• Equate MRs to MC
• R1 = P1 * Q1; R2 = P2 *Q2 • Π = R1 + R2 –
C
• Market with relatively inelastic demand
pays more

Third degree price discrimination
1/30/2022 372
 The goal of a monopoly—like that of any firm—is to earn the highest
profit possible. And, like other firms, a monopolist faces constraints.
Reread that last sentence because it is important. It is tempting to think
that a monopolist— because it faces no direct competitors in its market—
is free of constraints. Or that its constraints are special ones, unlike those
of any other firm. For example, many people think that the only force
preventing a monopolist from charging outrageously high prices is public
outrage. In this view, your cable company would charge $200, $500, or
even $10,000 per month if only it could “get away with it.” But with a little
reflection, it is easy to see that a monopolist faces purely economic
constraints that limit its behavior—constraints that are similar to those
faced by other, non-monopoly firms. What are these constraints?
 First, there is a constraint on the monopoly’s costs: For any level of output
the monopolist might produce, it must pay some total cost to produce it.
This cost constraint is determined by the monopolist’s production
technology—which tells it how much output it can produce with different
combinations of inputs —and also by the prices it must pay for those
inputs. In other words, the constraints on the monopolist’s costs are the
same as for any other type of firm, such as the perfectly competitive firm
we studied in the previous section.
 There is also a demand constraint. The monopolist’s demand curve—which
is also the market demand curve—tells us the maximum price the
monopolist can charge to sell any given quantity of output.

Monopoly and Dead Weight Loss
 Assume there is a tendency of moving from
competitive to monopoly output. If the
demand and total Cost functions are
Q=100-2P and TC=14Q+2Q2, respectively
◦ Determine Pc, Qc, Pm, and Qm.
◦ Show the equilibrium Q and P you obtained in A
above graphically.
◦ Calculate the CS and PS under competitive and
monopoly market structure.
◦ Calculate part of CS transferred to the
monopolist due to inefficiency of monopoly.
◦ Calculate the social cost (net loss or DWL) of
monopoly
Solution
Equilibrium Q and P in perfectly competitive market
P=MC
50-0.5Q=14+4Q
50-14=4Q+0.5Q
36=4.5Q
Qc=8
Pc=50-0.5Q or 14+4Q
=50-0.5(8) or 14+4(8)
=50-4 or 14+32
=46=46
Equilibrium Q and P in monopoly market
TR=PQ
=(50-0.5Q) Q
=50Q-0.5Q2
MR=∂TR=50-Q
∂Q
MC=∂TC=14+4Q
∂Q
MR=MC
50-Q=14+4Q
50-14=4Q+Q Dr Hassen B Market Structure and their
36=5Q 1/30/2022 optmization 375
 Qm=7.2
 Pm=50-0.5Q
 =50-0.5(7.2)
 =50-3.6 =46.4

MC=SS
Pm
Pc
DDm
MRm
Qm Qc

Consumers’ and producers’ surplus in perfect competition
C. CS=1/2 (50-46) x 8
=1/2(4) x 8
=16
PS=1/2(46-14) x 8
=1/2(32) x 8
=16 x 8=128
CS and PS under monopoly
CS =1/2(50-46.4) x 7.2
=1/2(3.6) x 7.2
=1.8 x 7.2=12.96
PS = ½((42.8-14) x (7.2))+((46.4-42.8) x 7.2)
=1/2((28.8) x (7.2))+(3.6 x 7.2)
=14.4(7.2)+ 25.92
=103.68+25.92=129.6
The CS transferred to the monopolist and DWL due to monopoly output
D. Consumers’ loss due to monopoly =CS under perfect –CS in monopoly
16-12.96
3.04
E. The amount of surplus transferred from consumers to the monopolist is
(Pm-Pc) x Qm
(46.4-46) x 7.2
0.4 x 7.2
2.88
DWL= CS not transferred + PS lost
=1/2(((Pm-Pc) x (Qc-Qm)) + ½ ((Pc-MC at
Qm) x (Qc-Qm)
=1/2(46.4-46) x (8-7.2) + ½((46-42.8) x (8-7.2)
=1/2 ((0.4) x (0.8)) + ½ ((3.2) x (0.8))
=0.2(0.8) + 1.6(0.8)
=0.16+1.28=1.44

OLIGOPOLY:
 In general, oligopoly market is divided in to two. These
are
 Non collusive oligopoly
 Collusive oligopoly. Following we will discuss models for oligopoly problems
under these two kinds of oligopoly.

 NON-COLLUSIVE OLIGOPOLY: This implies that
firms do not enter in to collusive agreement. There are
a number of non-collusive oligopoly models that give
us stable solution to the oligopoly problem that may
arise. Example
 The Cournot’s model (1838)
 The Kinked demand (Sweezy’s) model (1839)
 The Stackelberg’s model (1920)
 The Bertrand’s model (1883)
 The Chamberlain’s model (1883). For this course we will discuss only
some of them.

The Oligopoly Problem
 Principal features– Firms are few
and interdependent
– May produce similar or differentiated
products
◦ – Policies of one directly affects the other•
Must take its rivals’ action
◦ • Anticipate (rival plans to do, its reaction)
– Plan sequence of moves and set of
reactions
– Manager plays a game
Concepts of Game theory
 Playing a game requires:
– A set of players
– Alternative strategy
– A set of payoffs
 Zero sum game: one gains and the other losses

 Non-zero sum game:
– Non cooperative: do what is best for themselves
◦ Cooperative game: – collude – cartel
 Maximize joint profit and distribute
 No more than monopoly quantity
 Charge monopoly price
– Incentive to cheat: Prisoner’s Dilemma

Cheating and cooperative game
• Problem: contract enforcement

• Expanding increases profit
• Example:
• P = 20 – X
• Mc = 0
• Profit = 20X – x2
• X = 10, Price = 10
• Profit = 100, Profit (one)= 50, Profit
(two)=50

 Cheating Firm one– X1 = 6, X2 =5: X = 11
– P = 20 – 11 = 9
– Profit (one) = 6 * 9 = 54
– Profit (two) = 5 * 9 = 45
– Profit = 99
 Firm two does the same– X1 = 6, x2 = 6:X = 12

– P = 20 – 12 = 8
– Profit (one) = 6 * 8 = 48
– Profit (two) = 6 * 8 = 48
– Profit = 96

Cournot and Duopoly
 Augustine Cournot (1838) assumed that there are only two firms each having (owning) a
mineral water well and operating at zero cost. Let one of the firms A and the other B. Both
sell their output in a market with a down ward sloping DD curve. Each firm acts on the
assumption that its competitors will not change its output and decides its own output so as
to maximize profit.
 Mathematical version of the Cournot model. It is based

on the following assumptions
◦ Each firm maximizes profit by assuming the output of the other is
constant (ignoring their interdependence or naïve assumption).
◦ The duopolies face the same demand function (curve).
◦ The MRs of the duopolies need not be the same. This is because if the
duopolies are of unequal size, the one with the larger output or smaller
MC will have smaller MR. This implies that in the short run the
duopolies will supply different output levels but sell at the same price
since they supply identical products it is the total output that
determines price.
◦ The duopolies have different cost function.
◦ In the long run, however, each firm will supply 1/3 of the market
Example
 Assume that the market demand and cost functions of the
duopolies are
 P =100 - 0.5Q, where Q = q1+q2
 TC1= 5q1
 TC2 = 0.5q22.
 Given these answer the questions that follow
A. Determine the short run equilibrium output of each duopoly ignoring
their interdependence (with naive assumption)
B. What is the short run market price?
C. Find the demand functions of the duopolies (the reaction curves or
graphic solution of Cournot’ model and draw) and show the short run
output levels.
D. Calculate the short run profits of each duopoly and the industry profit.
E. Verify the economic profit of each duopoly graphically
F. Explain the relationship between output and MR in the short run.
G. Calculate the long run equilibrium output of each duopoly, market price,
and economic profits of each firm and the industry profit as a whole

Solution A. Short run price and quantity of market
1st find TR1 = Pq1
= (100 – 0.5 (q1+q2)) q1

= 100q1 –0.5q12 – 0.5q1q2
2nd find MR1 = ∂TR1 = 100 –q1 – 0.5q2
∂q1
3rd find MC1 = ∂TC1 = 5
∂q1
4th equate MR1 = MC1
100 – q1 – 0.5q2 = 5
100 – 5 - q1 – 0.5q2 = 0
95 – q1 – 0.5q2 = 0
95 = q1 + 0.5q2 -------------------------------------------- (1)
5th find TR2 = Pq2
= (100 – 0.5 (q1+q2)) q2
= 100q2 – 0.5q22 – 0.5q2q1
6th find MR2 = ∂TR2 = 100 – q2 – 0.5q1
∂q2
7th find MC2 = ∂TC2 = q2
∂q2
8th equate MR2 = MC2
100 – q2 – 0.5q1 = q2
100 – q2 – q2 – 0.5q1 = 0
100 – 2q2 – 0.5q1 = 0
100 = 2q2 +0.5q1 --------------------------------------- (2)
9th The profit maximizing (loss minimizing) output of q 1 and q2 can be solved from the two equations using simultaneous equation method. That is
q1 + 0.5q2 = 95 (0.5q1 + 2q2 = 100) (–2)
q1 + 0.5q2 = 95

optmization 387
1/30/2022
-q1 – 4q2 = -200
-3.5q2 = -105
q2 = 105/3.5 = 30, substituting this in any on of the above equation gives the value of
q1. That is
q1 + 0.5q2 = 95
q1 + 0.5 (30) = 95
q1 = 95 – 15 = 80
Q = q1 + q2 = 80 +30 = 110
B. Market price: P = 100 – 0.5Q, where q1 + q2
= 100 – 0.5 (80 + 30)
= 100 – 0.5 (110)
= 100 – 55 = 45
C. The demand functions (reaction curves) of the duopolies are obtained by solving
for q1 and q2 from the two equations as follows.
95 = q1 + 0.5q2
q1 = 95 – 0.5q2, is the demand function for firm 1. Hence,
If q2 = 0, then q1 = 95 and if q1 = 0, then q2 = 190
100 = 2q2 + 0.5q1
2q2 = 100 – 0.5q1
q2 = 50 – 0.25q1, is the demand function for firm 2. Hence

If q1 = 0, then q2 = 50 and if q2 = 0, then q1 = 200. The reaction
curves (graphic solution of Cournot’s model) is
q2
190 Firm 1’s reaction curve
50 Equilibrium
Firm 2’s reaction curve

30
q1
80 95 200
D. At the equilibrium each firm maximizes their own profit. But the industry profit is
not maximized. Why firms choose these sub optimal output? The reason is that,
the Cournot pattern of behaviour implies that the firms do not learn from past
experience, each expects the other to remain at a given position. Each firm acts
independently. That is, each does not know (recognise) the other will behave
(hold) the same assumption.
The economic profits of each duopoly
Π1 = Pq1 – TC1 Π2 = Pq2 – TC2
= 45(80) – 5(80) = 45 (30) – 0.5 (30) 2
= 3600 – 400 = 3200 =1350 – 450 = 900
Π= 3200 + 900 = 4100 is the total industry profit due to naïve assumption
E. The relevant curves to show profits graphically are
-The market DD curve
-The MR curve derived from the market DD curve
-Each firm’s MC and ATC curves
P
MC2
100 MC1 100 ATC2

ATC1
45
45
30
80 100 200 30 100 200

15
5
Π1 = q1 (P – ATC1) Π2 = q2 (P – ATC2)
= 80 (45 – 5) = 30 (45 – 15)
= 80 (40) = 30 (30)
= 3200 = 900

F. Firm 1 has lower MR than firm 2 because q1 > q2 (80 > 30)
MR1 = 100 – q1 – 0.5q2 MR2 = 100 – q2 – 0.5q1
= 100 – 80 – 0.5 (30) = 100 – 30 – 0.5(80)
= 100 – 80 – 15 = 100 – 30 – 40
=5 = 30
MR1 < MR2 because MC1 < MC2 and q1 >q2
G.The long run equilibrium output and price are calculated from the MR functions using simultaneous equation method. That is
q1 + 0.5q2 = 100
(0.5q1 + q2 = 100) (-2)
q1 + 0.5q2 = 100
-q1 – 2q2 = -200
-1.5q2 = -100
q2 = 100/1.5 = 66.7, substituting this in any of the above MR will give us q1 in the long run. That is
q1 + 0.5q2 = 100
q1 = 100 –0.5 (66.7)
q1 = 100 – 33.3 = 66.7
=> P = 100 – 0.5Q
P = 100 – 0.5 (66.7 +66.7)
P = 100 – 66.7 = 33.3
=> Π1 = Pq1 – TC1 Π2 = Pq2 – TC2
= 33.3 (66.7) – 5(66.7) = 33.3 (66.7) – 0.5 (66.7) 2
= 2221.11 – 333.5 = 2221.11 – 2224.45
= 1887.61 = -3.34
Π = Π1 + Π2 = 1887.61 + (-3.34) = 1884.27 is total industry profit in the long run. Note that, each firm’s and industry profits are higher in
the short run than in the long run.

THE STACKELBERG MODEL (Quantity
leadership)
 This model often used to describe industries in which there is a
dominant firm or a natural leader. For example, IBM is often considered
to be a dominant firm in the computer industry. A commonly observed
pattern of behaviour is for the smaller firms in the computer industry
to wait for IBM’s announcements of new products and then adjust their
own product decisions accordingly. In this case we might want to
model the computer industry with IBM playing the role of a Stackelberg
leader and the other firms in the industry being Stackelberg follower.
 Suppose that firm 1 is the leader and that it chooses to produce q1.
Firm 2 responds by choosing a quantity q2. Each firm knows that the
equilibrium price in the market depends on the total output produced.
That is by substituting Q (q1 +q2) in the inverse demand function
(curve).
 What output should the leader choose to produce to maximize profits?
The answer depends on how the leader thinks the followers will react
to its choice. Presumably, the leader should expect that the follower
will also attempt to maximize profits as well, given the choice made by
the leader. In order for the leader to make a sensible decision about its
own product, it has to consider the follower’s profit maximization
problem as its own.

Numerical example:
 Consider the example we have used to
describe Cournot’s model. That is, P =
100 – 0.5Q, where Q=q1 + q2, TC1 = 5q1,
and TC2 = 0.5q22. Given this,
 Find the equilibrium q1, q2, market price, Π1, and Π2
 Firm 1 being Stackelbrg’s sophisticated leader and firm 2 the
follower
 Firm 2 being Stacklberg’s sophisticated leader and 1 the
follower
 From the view point of profit obtained is it better
for the firms to be a leader or a follower?
 The reaction (DD) functions or curves are found by taking the partial
derivatives w.r.t. q1 and q2 and equating to zero.
 Π1= Pq1 – TC1= (100 –0.5 (q1+q2)) q1 –5q1

 = 100q1 – 0.5q12 – 0.5q1q2 – 5q1
 = 95q1 – 0.5q12 - 0.5q1q2
 Π2 = Pq2 – TC2 = (100 – 0.5 (q1+q2) q2 –0.5q22
 = 100q2 – 0.5q1q2 – 0.5q22 – 0.5q22
 = 100q2 – 0.5q1q2 – q22
◦ The partial derivatives w.r.t. q1 and q2
◦ ∂ Π1= 95 – 0.5q2 – q1
◦ ∂q1
◦ ∂ Π1= 100 – 0.5q1 – 2q2
◦ ∂q1
◦ The reaction (DD) function are
◦ q1= 95 – 0.5q2 --- firm 1 reaction (DD) function
◦ q2= 50 – 0.25q1 --- firm 2 reaction (DD) function
◦ Stakelberg’s solution with firm 1 being the sophisticated leader. Firm 1 will substitute firm 2’s reaction (DD) function in its own
profit equation to produce an output that will maximize profit as if it were a monopoly. That is
 Π1= Pq1 – TC1
 = 95q1 – 0.5q12 – 0.5q1q2, substituting firm 2’s DD function
 = 95q1 – 0.5q12 – 0.5q1 (50 – 0.25q1)
 = 95q1 – 0.5q12 – 25q1 + 0.125q12
 = 70q1 – 0.375q12
◦ The first order condition of the profit function w.r.t. q1
◦ ∂Π1= 70 – 0.75q1
 ∂q1
 = 70= 0.75q1
 = q1 = 70/0.75 = 93.333
 Π1= 70q1 – 0.375q12
 = 70 (93.333) – 0.375 (93.333) 2
 = 6533.333 – 3266.666 = 3266.66
◦ Firm 2 will substitute firm 1’s output in its own DD as a follower. That is
◦ q2 = 50 – 0.25q1
 = 50 – 0.25 (93.333)
 = 50 – 23.333 = 26.666
 Π2 = 100q2 – q22 – 0.5q1q2
 = 100 (26.666) – 26.6662 – 0.5 (93.333) (26.666)
 = 2666.66 – 711.1 – 0.5 (2488.8)
 = 2666.7 – 711.1 – 1244.4 = 711.1
◦ P = 100 – 0.5 Q
 = 100 – 0.5 (93.33 + 26.666) Dr Hassen B Market Structure and their
 = 100 – 0.5 (120) 1/30/2022 optmization 395
◦ Stakelberg’s solution with firm 2 being the sophisticated leader. It will substitute firm 1’s DD function in its own profit function to
produce an output that will maximize its profit as it were a monopoly. That is
◦ Π2 =Pq2 – TC2
◦ Π2 = 100q2 – q22 – 0.5q1q2, substituting firm 1’s DD function
 = 100q2 – q22 – 0.5q2 (95 – 0.5q2)
 = 100q2 – q22 – 47.5q2 + 0.25 q22
 = 52.5q2 – 0.75q22
◦ The first order condition of Π2 w.r.t.q2 gives
◦ ∂ Π2 = 52.5 – 1.5q2
 ∂q2
 = 52.5 = 1.5q2
 = q2 = 52.5/1.5 = 35
 Π2 = 52.2q2 – 0.75q22
 = 52.2 (35) – 0.75 (35) 2
 = 1837.5 – 0.75 (1225)
 = 1837.5 – 918.75 = 918.75
◦ As a follower firm 1 will substitute the output produced by firm 2 on its DD function. That is
◦ q1 = 95 – 0.5 q2
 = 95 – 0.5 (35)
 = 95 – 17.5 = 77.5
◦ Π1 = 95q1 – 0.5q12 – 0.5q1q2
 = 95 (77.5) – 0.5 (77.5) 2 – 0.5 (35) (77.5)
 = 7362.5 – 3003.125 –1356.25
 = 3003.125
◦ P = 100 – 0.5 (35 + 77.5)
 = 100 – 0.5 (112.5) 1/30/2022 optmization 396
THE BERTRAND’S MODEL
 (simultaneous price setting): This model assumed a model of competitive bidding and hence
is the opposite of the Cournot’ model.
◦ The Cournot’s model described that firms were choosing their quantities and letting the market
determines the price. Another approach is to think of firms as setting their prices and letting the
market determines the quantity sold. This model is known as the Bertrand model.
◦ What does the Bertrand model looks like? The answer is that when firms are selling identical
(homogenous) products and have significant effect on the price, the Bertrand equilibrium is a
competitive equilibrium for they engaged in strategic interaction. That is the Bertrand equilibrium is
where price equals MC. How?
◦ First, we note that price can never be less than MC. As a result, either firm would increase its profits
by producing less output. So let us consider the case where P >MC. Suppose that both firms are selling
at some >MC. Consider the position of firm 1. If it lowers its price by any small amount ε and if the
other firm keeps it price at , all the consumers will prefer to purchase from firm 1. By cutting its price
by an arbitrary small amount, form 1 can steel all the consumers from firm 2.
◦ If firm 1 really believes that firm 2 will charge a price that is greater than MC, it will always pay firm 1
to cut its price to - ε. But firm 2 can reason the same way. Thus, any price higher than MC cannot be
equilibrium. The only equilibrium is then the competitive equilibrium.
◦ This result seems paradoxical when you first encounter it. You may wonder how can we get a
competitive price if there are only two firms that produce identical products in the market? If we think
of the Bertrand model as a model of competitive bidding it makes more sense. Suppose that one firm
“bids” for the consumers’ business by quoting a price above MC.Then the other firm can always make
a profit by undercutting this price with a lower price. It follows that the only price that each firm can
not rationally expects to be undercut is a price equal to MC. Thus, it is often observed that
competitive bidding among firms that are unable to collude can result in prices that are much lower
than it can be achieved by other means. This phenomenon is simply an example
Dr Hassen of Bertrand
B Market Structure and their
competition. 1/30/2022 optmization 397
 Numerical Example: Given P = 100 – 0.5Q, where Q = q1+q2, TC1= 5q1, TC2 = 0.5q22 find the Bertrand’s equilibrium.
 Solution:
 Firm 1 Firm 2
 P = MC1 P = MC2
 P=5 P = q2
 5 = 100 – 0.5Q
 -95 = -0.5Q
 Q = 95/0.5 = 190
◦ Check that P = MC1 = MC2
 100 – 0.5 (q1 + q2) = 5 100 – 0.5 (q1 + q2) = q2
 95 – 0.5q1 – 0.5q2 100 – 0.5q1 –1.5q2
 95 = 0.5q1 + 0.5q2 100 = 0.5q1 + 1.5q2
◦ Solving the two equations simultaneously
◦ 95 = 0.5q1 + 0.5q2
 (100 = 0.5q1 +1.5q2) –1
 95 = 0.5q1 + 0.5q2
 -100 = -0.5q1 – 1.5q2
 -5 = -q2, q2 = 5
 95 = 0.5q1 + 0.5 (5)
 95 = 0.5q1 + 2.5
 0.5q1 = 95 – 2.5
 0.5q1 = 92.5, q1 = 92.5/0.5 = 185 and Q = 5+185 = 190
◦ P = 100 – 0.5Q
 = 100 – 0.5 (5 + 185) 1/30/2022 optmization 398
The End
Stay Safe
Nagaadhaan
399

Chari Boru's - DR - Haasan Huseen - Microeconomics I PPT (PHD) Program

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chari Boru's - DR - Haasan Huseen - Microeconomics I PPT (PHD) Program

Uploaded by

Copyright:

Available Formats

Bulehora University

Associate professor in Applied and Agricultural

 While most economic models are

 Distinction between positive and

 Ceteris Paribus means “other things the same”

 Optimization models appear to be

 Normative economic theories focus on

The demand curve has a

Q* Quantity per period

 A more general model is

A shift in demand will lead to a new equilibrium:

Price An increase in demand...

500 750 Quantity per period

 when x=5, y=13.2, the slope= -2(5)/13.2= -0.76

 the slope rises as y rises

 Clarification of the basic behavioral

 Economics is the study of how scarce

 The most commonly used economic

 The supply-demand model is only a

 Testing the validity of a model is a

 The basic economic model o of consumer

 For simplicity, two goods assumption (one

 If the consumer prefers one bundle to

 Completeness: For all x and y, in X, either x ≽ y, or x ≼ y,

◦ if A and B are any two situations, an individual can always specify

 Strict convexity: given x ≠ y and z in X, if x ≿ z, and y ≿ z, then tx + (1-t)y

 Convexity implies an economic agent prefers averages to extremes. Convex

 There are several equivalent ways of

 Given that preferences are complete, reflexive,

 Economists have abandoned the old-fashioned view of

i.e. More desirable situation offer more utility than less

if A is preferred to B, then the

 The set of all bundles on or above an

 Diminishing MRS: The MRS of x for y, diminishes as x

• Solving for MRS = -dy/dx:

• Along any indifference curve, utility is

• For this example, it is easier to use the

 The utility maximization model predicts behavior even

 Utility maximization does not imply “selfish” behavior.

 In order to maximize utility, given a fixed amount of

 Nothing in the utility-maximization model prevents

◦ The third issue is that if we multiply all prices and

 The left hand side fraction shows the MRS between

 Both the primary utility maximization problem and the

 Because of individual’s desire to maximize utility, given a budget constraint, the

Example, Nicholson, page 116, example 4.5.

 Properties of the indirect utility function

 Properties of expenditure function

 Hicksian demand function:

1. e(p, v(p,m)) ≡ m. The minimum expenditure necessary to reach utility v(p,

 Example:Varian (1992), page 111-112, The Cobb-

 Example:Varian (1992), page 112, The CES utility

 Exercises: Nicholson, page 133-124, problems 4.1, 4.3,

 Extensions: Nicholson, page 126 E4.1, E4.2, and E4.3

 An important feature of the direct and indirect compensation

 Such functions are very useful in welfare economics.

 To maximize utility, given a fixed amount of income to

 Assume that the individual’s MRS = 1

Quantity of y The individual can afford

If all income is spent

 The tangency rule is only necessary but not sufficient unless

 The individual’s objective is to maximize

 Set up the Lagrangian:

At p1, the person spends E(p1,…)