Topic One: Theory of Consumer Behaviour

We know that market demand and supply curves leads to market equilibrium. We also know that
these curves are obtained by horizontal summation of the individual demand and supply curves.
However, we not aware that there is equilibrium of an individual consumer and an individual
producer who are responsible for individual demand and individual supply curves. In this topic
we learn about equilibrium of the individual consumer and in the second topic we will focus on
that of the individual producer.

There are three approaches to analysis of demand: cardinal utility approach; indifference
curves/ordinal utility approach, and revealed preference approach.

Cardinal utility approach

Utility is ability of a good to satisfy a want. Concept of utility originated from Jeremy Bentham
and his fellow utilitarians (such as John Stuart Mills) of 18th century. The concept acquired its
precise meaning in 19 th century through collective efforts of William Stanley Jevons; Leon
Waras; Carl Menger and Alfred Marshall. Cardinal utility approach assigns a value to the level
of satisfaction associated with the consumption of a basket of goods: actual measure of utility, in
util. For instance, we can say an individual receive 30 utils from consuming one beer, he/she
receives 50 utils, or 20 additional utils, from consuming two beers. That is, it provides an actual
index or measure of satisfaction.
Assumptions of cardinal utility approach
i) Consumers are rational: The consumer is a rational person aiming at maximizing total
utility. The maximization is subject to constraint imposed by his income, which is fixed. A
rational consumer attempts to attain the highest level of utility as possible with a given level of
income.
ii) Utility is cardinal: utility of a commodity is numerically measurable (utils vs money- where
util is an arbitrary unit of measurement of utility). Amount of money a consumer is prepared to
pay for a unit of commodity is most convenient unit of measurement.
iii) Marginal utility of money is constant: at different levels of consumer’s income; for rich
and poor.
iv) Diminishing marginal utility: The extra utility received from consuming one additional unit
of a good/commodity has a diminishing trend. That is; marginal utility of a commodity

1
diminishes each time a consumer has an additional unit of the commodity for his/her
consumption.
v) Additivity of utility: The total utility of a basket of goods is the sum-total of the utilities of
the individual commodities in the basket.

If there are n commodities in the basket with quantities x1 , x 2 ,...., x n , the total utility is given as:

U  f ( x1 , x2 ,..., xn )  U1 ( x1 )  U 2 ( x2 )  ....  U n ( xn )

Utility function
It is a table, a graph, or an equation showing the maximum satisfaction that can be achieved from
consuming any specified combination of goods. It is a way of assigning numbers to every
consumption bundle so that more preferable bundles get assigned a larger number than a less
preferred bundle (a bundle is a particular combination of two or more goods). That is, a bundle
( x1 , y1 ) is preferred over bundle ( x2 , y2 ) if and only if the utility of bundle ( x1 , y1 ) is larger than
the utility of ( x2 , y2 ) ; symbolically, ( x1 , y1 )  ( x2 , y2 ) if and only if u( x1 , y1 )  u( x2 , y2 ).

So a utility function can be written as: U  f ( X , Y ) where U denote utility, X is good x and Y
is good y .
Beers consumed Total utility Marginal utility
0 0 -
1 30 30
2 50 20
3 60 10
4 62 2
5 60 -2

Thus, total utility is the sum of the satisfaction derived from the consumption of several goods.
Marginal utility is the increase in total utility following the consumption of an extra unit of a
good. Marginal utility of a good MU measures the increase (or decrease) in total utility (U )
U
following a small variation in the quantity consumed (U ) : MU 
x

2
Thus: marginal utility is the first derivative of the utility function. It gives the slope of the utility
function. The marginal utility of the good (beer) gets smaller as the quantity consumed increases.
This phenomenon is called the law of diminishing marginal utility.

Consumer’s equilibrium
In case of a single commodity, the consumer will attain his/her equilibrium when the marginal
utility of X is equated to the price of the commodity. That is if utility is cardinally measurable,
the condition for constrained utility maximization would be for the consumer to spend all income
on X in such a way that:
MU X  PX
Thus, marginal utility of good X equals the price of good. If MU X  PX , the consumer can

increase his/her satisfaction by having more of X ; If MU X  PX , the consumer can increase

his/her satisfaction by reducing consumption and holding his/her income unspent.
Number of beers consumed Marginal utility Price ($)
1 8 2
2 7 2
3 6 2
4 5 2
5 4 2
6 3 2
7 2 2
8 1 2
9 0 2
10 -1 2
11 -2 2
12 -3 2

The equilibrium level of consumption of beers is seven as the marginal utility of the seventh beer
equals its price.

3
In a multiple commodity case, the condition for consumer’s equilibrium is expressed as:
MU X MU Y MU n
  .....
PX PY Pn
That is, the consumer will be in equilibrium when the ratios of the marginal utilities of the
individual commodities to their market prices are the same. Note: MU X / PX is the extra or
marginal utility per dollar/shilling spent on X, thus for constrained utility
maximization/optimization, the marginal utility of the last dollar/shilling spent on X , Y ,...., n
should be the same.

A hypothetical example: A consumer wishes to distribute expenditure of $ 44 between goods X ,

Y and Z . Px  $8, Py  $4 , and Pz  $2

Units consumed MU x MU y MU z MU x / Px MU y / Py MU z / Pz

1 72 60 64 9 15 32
2 48 44 56 6 11 28
3 40 32 40 5 8 20
4 36 24 28 4.5 6 14
5 32 20 16 4 5 8
6 20 8 12 2.5 2 6
7 12 4 8 1.5 1 4

In order to maximize utility, the consumer must allocate available income so that:
MU X MU Y MU Z
 
PX PY PZ
This yields a selection where the consumer buys 2 units of X , 4 units of Y and 6 units of Z . The
total income spent is given by 2 units of X @ $ 8  4 units of Y @ $ 4  6 units of Z @ $2  $44
48 24 12
 
8 4 2
Criticism of the cardinal utility approach
i) The assumption of cardinal utility is unrealistic: utility is an abstract attribute like honesty,
intelligence etc. Numerical measurement of such attributes is not possible.

4
ii) The assumption of constant marginal utility of money is unrealistic. Marginal utility of
money increases with increasing expenditure. Also, marginal utility of money is different at
different levels of income.
iii) The law of diminishing marginal of commodity may not hold. Marginal utility of a
commodity only holds for certain goods and for certain circumstances. Marginal utility of
commodities such as money, diamonds, never diminishes the same for many other durable and
semi-durable goods. Also, marginal utility of a commodity diminishes only when all units of a
commodity are consumed at a point of time, one after another; otherwise if a consumer consumes
one unit of commodity every day, the marginal utility may be same every day.

These drawbacks were solved by the introduction of ordinal utility: it’s more general, more
realistic and more powerful.

Ordinal utility/Indifference curve approach

This approach to analysis of demand was developed by English economist Francis Ysidro
Edgeworth (1881), extended by Italian economist Vilfredo Pareto (1906) and Soviet economist
Eugen Slutsky (1915), and was given a logical perfection by two English economists: Roy
George Douglas Allen and John Richard Hicks (1934). Hicks further developed it in his books
“Value and Capital” (1939) and “Revision in Demand Theory” (1956). This approach only ranks
the utility received from the utility received from consuming various amounts of a good/basket
of goods. Ordinal utility specifies that consuming two beers gives the individual more utility than
consuming one beer, but does not specify exactly how much additional utility the second beer
provides. i.e ordinal utility only ranks various consumption bundles.

Assumptions of ordinal utility approach

i) Rationality of the consumer: Consumer is assumed to be a rational person, always aiming at
maximizing his/her utility under given level of income and prices of the goods. The consumer is
also assumed to have all the relevant information to serve his objectives.
ii) Utility is ordinal: The consumer is capable of ranking different baskets of goods in order of
his preferences.
iii) Diminishing marginal rate of substitution: The rate at which a consumer foregoes
quantities of one good, say Y for the sake of an additional unit of the other, say X shows a

5
diminishing trend as more and more of the latter ( X ) are had. Thus, the slope of the indifference
curve increases from left to right as we go down the curve. Marginal rate of substitution of X for
Y ( MRS X ,Y ) , representing the numerical value of the slope of the indifference curve, thus,

diminishes with successive additional units of good X a consumer has in place of Y .

iv) Total utility of the consumer depends on the quantities of commodities consumed: if
x1 , x 2 ,...., xn represent quantities of goods X 1 , X 2 ,....., X n ; total utility U is given by:

U  f ( x1 , x2 ,...., xn )

 u1 ( x1 )  u 2 ( x2 )  .....  u n ( xn )
v) Consistency and Transitivity of choice: The consumer is assumed to be consistent in his/her
choices: if in one period he chooses bundle A over bundle B , he/she will not choose bundle B
over bundle A in another period, when both the bundles are available to him. Symbolically:
If A  B , then B  A (note the second part should read B is not preferred to A)
Similarly, his/her choices are characterized by transitivity: if A is preferred to B and B to C , then
A is preferred to C . Symbolically:
If A  B and if B  C , then A  C

Properties of preference ordering

i) Completeness: the consumer is able to rank all possible combinations of goods and services. If
A and B are bundles of goods then: either A is preferred to B , or B is preferred to A , or A and B
are indifferent.
ii)Transitivity: for any three bundles A, B and C , if A is preferred to B and B to C , then A is
preferred to C .
iii) Continuity: if A is preferred to B and C is close to B , then A is preferred to C
iv) Monotonicity/Non-satiation/ more-is-better: other things equal, more of a good is preferred
to less. That is, a consumer prefers having more of either good to having less. If a consumer
prefers A to be B ( A  B ) and also prefers C to D ( C  D ), then, this consumer prefers ( A, D) to
( B, D) , and also prefers ( B, C ) to ( B, D)

6
Note: non-satiation implies that points above an indifference curve are preferred to points on the
indifference curve. The points in the shaded region are in the preferred set.
v) Convexity: mixture of goods is preferable to extremes/averages of goods are better than
extremes. Convexity implies preference for averages. (NB: for a convex function, any line
joining two points on its graph lies entirely on or above the graph):

In the above figure, C is the average of A and B . Convexity implies that C is preferred to A and
B.
Note: the assumption of completeness, transitivity and continuity implies that preferences can be
represented by a utility function and indifference curves. Non-satiation implies downward
sloping indifference curves. Convexity implies that marginal rate of substitution of X for Y falls
as more of X is substituted for Y : indifference curves are convex to the origin when principle of
diminishing marginal substitution holds.

7
Indifference curves
An indifference curve is a graphical representation of a utility function. It is a locus of various

combinations of quantities of two goods each of which offers the same level of satisfaction (U )
to the consumer to the extent that the consumer turns indifferent in between any two of them.
Assume that a consumer in a two-commodity world: he/she consumes quantities x and y of
goods X and Y respectively. Total utility derived from these quantities is given by:
U  f ( x, y )

To derive a fixed level of utility, U a consumer may have more of X with less of Y , or more of
Y with less of X . Level of utility derived from each being the same, the consumer will be
indifferent in between the two: i.e it would not matter whether a consumer has more of X with
less of Y or more of Y with less of X so long as he has the same level of utility. Demonstration:
let the two commodities be oranges X and apples Y .

As long as the utility derived by the consumer from 10 oranges with 50 apples is the same as that
derived from 50 oranges with 10 apples: for the consumer each bundle is as good as itself.
Plotting the combinations of the two fruits with apple on the y-axis and orange on the x-axis and
joining the points by a smooth curve we get a curve referred to as the indifference curve.

8
Oranges Apples Change Change Slope of IC MRS X ,Y
X Y x y y / x | y / x |

10 50 - - - -
20 30 -20 10 -2 2
30 20 -10 10 -1 1
40 14 -6 10 -0.6 0.6
50 10 -4 10 -0.4 0.4

The numerical value of the slope ( MRS X ,Y ) decreases as we go down the curve. This is referred

to as the law of diminishing marginal rate of substitution of X for Y Consumer preferences such
as these that lead to an indifference curve that is convex to the origin or to the diminishing
marginal rate of commodity substitution are known as the convex preferences. Such preferences
lead to a negative slope of the indifference curve and are referred to as monotonic preferences.
So, what is a monotonic preference? If x1 , y1  is a bundle of two goods and if x2 , y 2  is a
bundle of two goods with at least as much of the both goods and one more of one, consumer’s
preference to x2 , y 2  over x1 , y1  expressed as x2 , y2   x1 , y1  is called monotonic
preferences. We further explain the concept of monotonic preferences using the following figure.

Bundle ( x1 , y1 ) lies on the indifference curve. Bundles with at least as much of both goods and
more of one, if preferred by a consumer, would imply his monotonic preferences. Starting from
bundle ( x1 , y1 ) and moving up anywhere up to the right is moving to the preferred zone while

9
moving down to the left is moving to non-preferred zone. Indifference curve is therefore the
lower boundary of the preferred zone and this is possible only when the curve is negatively
sloped.
Properties of indifference curves
i) Indifference curves are ubiquitous/everywhere dense: any bundle has an indifference curve
passing through it. That is, with the XY positive quadrant, every point represents a combination
of two goods X and Y that yield a certain level of utility to the consumer and there is an
indifference curve that pass through each of these combinations.
ii) Indifference curves are downward sloping from left to right: since we assume that the
consumer is rational, then if a consumer gives up some units of good Y , he/she will want more
units of good X , if he/she is to keep the level of utility unchanged. This suggests a
negative/inverse slope for the indifference curve. Otherwise, a positively sloped indifference
curve would imply that a consumer would have to give up units of both goods to retain any level
of utility. This scenario would be illogical because consumption of less of both goods cannot
leave total utility undiminished.
iii) Indifference curves cannot cross/do not intersect: it would be illogical for indifference
curves to cross since this would imply that a consumer is equally satisfied with the two
combinations of good X and Y , both of which have the same quantity of good X but different
quantities of good Y

Since combination A and C are on the same indifference curve, the consumer must be indifferent
between them. Combination B and C , however are on the same indifference curve, therefore the
consumer must be indifferent between them as well. This is however illogical since A contains

10
more of good Y and the same amount of good X as combination B , and so must be preferred to
it. This kind of illogical result occurs whenever indifference curve cross. Thus, indifference
curves should never cross each other.
iv) Indifference curves are convex to the origin: convexity ensures that as a consumer gives up
more units of one good say Y , then relatively more and more units of good X would be
consumed. For rationality, it would be rational to suppose that successive bigger quantities of
good X must be obtained to compensate the consumer for the loss of good Y and retain him/her
on the same level of utility. This proposition is referred to as diminishing marginal rate of
substitution.

Budget constraint
The amount of goods that a consumer can purchase is limited by the consumer’s income and by
the prices of the goods that he/she must pay. Thus, budget constraint is the limitation on the
amount of goods that a consumer can purchase imposed by his/her limited income and the prices
of the goods. It is the set of all bundles that exactly exhaust the consumer’s income at a given
prices. Its slope is the negative of the price ratio of the two goods. Budget line: it’s a line
showing various combinations of two goods that a consumer can purchase using all their income.
Assume a world with only two consumer goods: X and Y . PX is the price of good X . It is
measured in dollars/shilling per unit of good X . The consumer pays this price no matter what
quantity he/she buys. That is, there are no quantity discounts and there is no rationing. X is also
the quantity of good X that is purchased by the consumer. It is measured in units of good X per
unit of time. PX  X  PX X is the consumer’s expenditure on good X . Total expenditure

 PX X  PY Y . Let M is the consumer’s income/budget. The consumer cannot spend more than
his/her budget allows. The consumer’s budget constraint can be expressed as: PX X  PY Y  M

A rational consumer will spend every coin available. Thus, PX X  PY Y  M becomes

PX X  PY Y  M

11
Consumer equilibrium
A rational consumer seeks to maximize the utility or satisfaction received in spending his or her
income. A rational consumer maximizes utility by trying to attain the highest indifference curve
possible, given his or her budget line. This occurs where an indifference curve is tangent to the
budget line so that the slope of the indifference curve (MRS ) is equal to the slope of the budget

line ( PX / PY ) . The condition for constrained utility maximization/consumer optimization/

consumer equilibrium occurs where the consumer spends all income (i.e., he/she is on the budget
line) and
PX
MRS 
PY

12
Marginal Rate of Substitution
Shows the rate at which one good can be substituted for another good, if utility remains constant
(slope of the indifference curve)
Given U  f ( X , Y )
where U is utility,

MU X PX
MRS   
MU Y PY
Example
u( x1 , x2 )  x1a x2b
u ( x1 , x 2 ) / x1
MRS  
u ( x1 , x 2 ) / x 2
ax1a 1 x 2b

bx1a x 2b 1
ax1a 1 a x 2b b 1

b
ax
 2
bx1

Demand function
The consumer’s demand functions give the optimal amounts of each of the goods as a function
of the prices and income faced by the consumer. We write the demand functions as
x1  x1 ( p1 , p 2 , m)
x 2  x 2 ( p1 , p 2 , m)
The LHS of each equation stands for the quantity demanded. The RHS of each equation is the
function that relates the prices and income to that quantity.
Example
Using a Cobb-Douglas utility function of the form u  u ( x1 , x2 )  x10.5 x20.5 derive the demand

functions for each of the two goods (i.e x1 and x 2 ).

First we form the constrained utility maximization problem:
Maximize: u  x10.5 x20.5

Subject to: p1 x1  p2 x2  m

13
We form the Lagrangia2n function
L  x10.5 x20.5   (m  p1 x1  p2 x2 )
Apply the first order conditions. Equate to zero the first partial derivatives of L
u
Lx1   0.5 x10.5 x 20.5  p1  0...........................(i )
x1
u
L x2   0.5 x10.5 x 20.5  p 2  0...........................(ii )
x 2
u
L   m  p1 x1  p 2 x 2  0...............................(iii )

Divide equation (i) by equation (ii)
0.5 x10.5 x 20.5 p1
 ...........................(iv )
0.5 x10.5 x 20.5 p 2

x 20.50.5 p
0.5 0.5
 1 ...........................(v)
x1 p2

x2 p
 1 ...........................(vi)
x1 p2

 x1 p1  x2 p2 ...........................(vii)

x1 p1
 x2  ...........................(viii)
p2
Plugging equation (viii) into the budget constraint we get:

x p 
x1 p1   1 1  p2  m...........................(ix )
 p2 

We get:
x1 p1  x1 p1  m...........................( x)
 2 x1 p1  m
m
x1*   demand function for good one
2 p1
m
x 2*   demand function for good two
2 p2

14
Examples
i)A consumer spends whole of his income on two goods X and Y . When his income is $100,
the price of good X is $5 per unit and that of Y is $10 per unit. What can you infer about
MRS X ,Y at the optimum? In equilibrium MRS X ,Y   PX PY   5 10  0.5

ii)An individual’s marginal utilities for good X and Y are given as

MU X  40  5x, MU Y  20  3 y Determine MRS X ,Y when the consumption basket is

x  3, y  5. If PX PY  5 , is the basket an optimal one?

MRS X ,Y  5 1, Yes

iii)A consumer is found spending whole of his income Y . His income is $100 , price of X is $2
per unit and that of Y is $4 per unit. What can you say about MRS X ,Y at the optimum? Spending

the whole of his income on Y implies a corner solution with 0 units of X . In such case, the
indifference curve must have a lower slope than the budget line. MRS X ,Y would then be less than

the numerical value of the slope of the budget line, i.e. less than 1 2.

iv) Given the utility function

1 2

U  1000 x y
3 3

and the consumer budget

3000  100 x  50 y
Determine the equilibrium income for the consumer
Solution: U x  1000 3 *  y x  3  MU X
2

U y  2000 3 * x y 3  MU Y
1

(1000 / 3) ( y / x) 2 / 3 1 ( y ) 2 / 31 / 3
MU X MU Y  *  *
(2000 / 3) ( x / y )1 / 3 2 ( x)1 / 3 2 / 3
  y 2x

15
 MU X PX
At the equilibrium:  . Recall PX  100 and PY  50 . Therefore, we have
MU X PY

y 2x  100 50  y  4x . Substituting y  4 x in the equation of a budget line, we obtain the
equilibrium basket for the consumer: x  10, and y  40
v)How will the answer to above problem change if the utility functions were given as follows?
U  100 x  50 y
U  50 x  100 y
U  200 x  50 y

U x  100  MU X
U y  50  MU Y
Since PX  100 and PY  50 we have

MU X MU Y  PX PY  2
 MRS X ,Y  PX PY  2

Both budget line and utility function are coinciding/coincides. There are infinitely many
solutions i.e any basket on the budget line is as good as any other on it.
vi)Outline any four types of utility functions

Perfect Substitutes goods

U  X 1 , X 2   X 1  X 2
Preferences for perfect substitutes are represented by linear utility functions of the form. Perfect
substitutes have an additive utility function which implies that a consumer will yield some level
of utility even by consuming only one of the goods.
Complementary Goods
U  X 1 , X 2   Min X 1 X 2   MinX 1 , X 2 
These goods used together. The indifference curves are L-shaped.
Imperfect Substitutes
U  X 1 , X 2   X 1 X 2
Some level of utility is only possible when a combination of the goods is consumed, i.e. no
amount of utility is achievable from consuming zero amount of one good. Such goods are said to
be imperfect substitutes and their preferences take multiplicative form of utility functions. This

16
kind of a function is referred to as a Cobb-Douglas utility function. It yields convex monotonic
indifference curve/preferences and are said to be well behaved.
Quasi-linear utility function
Utility function is partly linear and partly non-linear.

U  X  Y or U  X  Y or U  log X  U or U  X  log Y

vii)How would you sketch the likely shape of the indifference curves in each of the following
cases?

a) When both goods are bad goods

b) When one of the good is a bad good
c) When one of the two good is a neutral good
d) When both of them are perfect substitutes
e) When both of them are perfect compliments
f) When both goods are imperfect substitutes

Income and substitution effect of a price change

The effect of a price fall of good X on the consumer equilibrium point is shown in the figure
below. The equilibrium point moves from point C to D , after the equilibrium pivots from AB to
AB' . The curves that join different consumer equilibrium points after a price change is called
price consumption curve or a price offer curve.

17
The fall in the price of good X causes the consumer’s demand to increase from X 1 units to X 2
units. This change in demand is referred to as the total effect of a price change. There are two
possible reasons for this;

As the price of good X falls, it becomes relatively cheaper while Y becomes relatively more
expensive. The consumer is therefore induced to substitute good X for good Y . This is referred
to as the substitution effect of a price change.

As the price of good X falls, the consumer experiences an increase in real income, i.e. his/her
purchasing power is increased because the income can now buy more of the good. This would
induce the consumer to buy more of good X (however, the consumer may also purchase more of
good Y with the extra income). This is referred to as the income effect of a price change.

The total effect of a price change (fall) indicated by the movement from point C to D , increasing
the demand of good X from X 1 to X 2 can be split into two effects: substitution effect plus the
income effect of a price change.

To separate the two effects, we assume that after the price fall, the consumer’s income is
adjusted so as to retain the consumer on the same level of utility. Since a price fall would
increase the consumer’s real income, a negative adjustment is made to the consumer’s real
income so that he/she has only enough to maintain him/her on the same level of utility as before
the price change. This negative adjustment on the consumer’s income after a price fall is referred
to as a compensating variation. (The positive adjustment in consumer’s income after a price rise
so as to maintain him/her on the same level of utility as before the price rise is called an
equivalent variation).

Graphically this is shown by a parallel shift of the new budget line AB until it is tangential to the
initial indifference curve U 1 . This is the budget line EF .

18
 Good Y A

E
e1
Y1 e2
e1'
Y2
U2
U1

O
Good X
X1 X1’ B X2 F B’
The movement from point e1 to e1' is the substitution effect. The consumer has substituted

X 1'  X 1 units of good X for Y1  Y2  units of good Y , so as to remain on the same level of utility

U1 .

However, the compensating variation is a device to help us to separate the substitution and
income effects of a price change. In reals sense, no real adjustment is made to the consumer’s
income and if we therefore assume at this point that the consumer’s extra income is given back
to him or her, then the budget line EF would shift to AB ' . The movement from point e1 to e2 is

the income effect of a price change thereby increasing the demand of good X further from X 1

to X 2 . The consumer buys more of good X because of the increase in real income.

NB: The substitution effect will always act in a way that when a price of a good falls ceteris
paribus, more of it will be purchased. The income effect however, can work either way: - when a
consumer’s real income rises, more or less of good X may be purchased. If more is bought, the
good is said to be a normal good while if less is bought, the good is said to be an inferior good. If
the fall in demand due to an increase in real income is larger than the increase in demand due to
the substitution effect, such a good is said to be too inferior and is called Giffen good.

19
Criticism of indifference curve approach
i)The theory is are based on unrealistic/strong assumption of rationality. A consumer may not
always act rationally.
ii) The theory overlooks important influences on consumer behavior. That is, it ignores important
influences of advertising, habits, customs, impulsiveness and speculative nature on consumer
behavior.
iii) The theory is based on the assumption of perfect competition and perfect knowledge of
prices.
iv)The approach is based on two goods only; consumers generally buy a large number of
commodities to satisfy their wants.
v) The approach retains some of the weaknesses of the cardinalist approach. Tells us nothing
new, it is similar to utility analysis i.e. diminishing marginal utility is replaced by diminishing
MRS. Conditions of equilibrium are similar. It substitutes the concept of preference for utility.

iv) Fails to explain how a consumer makes choices under risk and uncertainty.
v) It is based on hypothetical combinations, some absurd or not possible in real life.
vi) Not based on observed market behavior. Indifference curves are non-transitive: a consumer
may be indifferent because of his inability to judge the difference between alternative
combinations.
Revealed Preference Approach
Revealed preference introduced in 1938 by Paul. A. Samuelson “Consumption Theory in Terms
of Revealed Preference”. The term revealed preference also appeared in “Fundamental of
Economics Analysis” (1947). Revealed preference theory is behaviouristic rather than
introspective as in Hicks-Allen approach of indifference curve. The theory was designed to do
away with the subjectivity element that is implicit in utility theories of both marginal and the
indifference curves approach.

Choosing a basket of goods in a given budget situation means that a consumer has revealed his
preference for the basket. The rest are either more costly or inferior.

20
AB: budget line. Baskets D and E are affordable while F is not. If consumer chooses basket E >
he reveals preference for this basket. Other baskets are more costly (e.g F) or inferior like those
below it like D or all baskets other than E on AB. Choice of basket D violates assumption of
rationality, as it does not maximize a consumer’s utility. Axiomatically, the basket revealed
preferred must max the consumer utility.

Assumptions of revealed preference theory

i. Rationality: consumer is rational and prefers baskets that include more quantities of the
commodities.
ii. Consistency: consumer is consistent in his behaviour. If he chooses bundle A when
Bundle B is also available, he will not choose B if bundle A is also available to him. If
A  B, then B  A .

iii. Transitivity: in any particular situation, if A  B and B  C  A  C

iv. Revealed preference axiom: a consumer by choosing a basket of goods in a given
budgetary condition reveals his preference for the basket. The chosen bundle max
(axiomatically) his utility. The rest of the baskets on the Budget line are revealed inferior.

Combining preference ordering principles with the revealed- preference-axiom we get Weak and
Strong Axioms of revealed preference theory:

21
Weak Axiom of Revealed Preference (WARP)
If bundle x1 , y1  is directly revealed preferred to bundle x2 , y2  , the two bundles being
different from each other, it cannot happen that bundle x2 , y2  would be directly preferred to
bundle x1 , x2  .

i.e if bundle x1 , y1  is bought at prices  p1 , q1  and a different bundle x2 , y 2  is bought at
prices  p2 , q2  , then if;
x1 p1  y1q1  x2 p1  y2 q1
It must not be the case that
x2 p2  y2 q2  x1 p2  y1q2 instead
x2 p2  y2 q2  x1 p2  y1q2

Simply, it means that if the second bundle x2 , y 2  is affordable when first bundle x1 , y1  is
bought, when the second bundle is bought, the first bundle must not be affordable.

Above 3 equations can be re-written as;

If x2 p2  y2 q2  x1 p2  y1q2
It must not be the case that
x1 p1  y1q1  x2 p1  y2 q1 instead
x1 p1  y1q1  x2 p1  y2 q1

Simply, it means that if the first bundle ( x1 , y1 ) is affordable when second bundle x2 , y 2  is
bought, when the first bundle is bought, the second bundle must not be affordable.
Example: six apples and four oranges are bought when prices of the two fruits are one dollar per
unit each. When price of an apple rises to $.2 per unit and that of an orange falls to $0.5 per unit,
the quantities of the two fruits that are bought are 8 apples and 3 oranges. Do these observations
indicate violation of WARP?
x1  6, y1  4; p1  1, q1  1
x2  8, y2  3; p2  2, q2  0.5

According to condition (iv)

x2 p2  y2 q2  8  2  3  0.5  17.5
x1 p2  y1q2  6  2  4  0.5  14
17.5  14

22
 x2 p2  y2 q2  x1 p2  y1q2

If x2 p2  y2 q2  x1 p2  y1q2
It must not be the case that
x1 p1  y1q1  x2 p1  y2 q1 instead
x1 p1  y1q1  x2 p1  y2 q1
x1 p1  y1q1  6 1  4 1  10
x2 p1  y2 q1  8 1  3 1  11
10  11

 x1 p1  y1q1  x2 p1  y2 q1
 x1 p1  y1q1  x2 p1  y2 q1 ….
 WARP is not violated & the buying behaviour of the consumer is consistent.

Bundles 1(6,4) 2(8,3)

Price sets
1(1,1) 10 11
2(2,0.5) 14* 17.5

When bundle 2 (row 2, col 2) is bought, bundle 1 (row 2, col 1) is affordable and when bundle 1
(row 1, col 1) is bought, bundle 2 (row 1, col 2) is not affordable as required by WARP. Thus,
choices are consistent.
Question
In the above illustration, let
x1  2, y1  4; p1  2, q1  4
x2  4, y2  2; p2  4, q2  2
Investigate whether WARP is violated or not

Strong Axiom of Revealed Preference

If x1 , y1  is revealed preferred to x2 , y 2  (directly/indirectly), the 2 bundles being different

from each other, it can’t happen that x2 , y 2  would be revealed preferred to x1 , y1  directly or
indirectly.

WARP>>if bundle A is revealed preferred directly to bundle B , we should never observe B

being directly revealed preferred to A.
If bundle A is revealed preferred to bundle C & bundle C is revealed preferred to bundle B ,

23
>then bundle A is revealed preferred to bundle B indirectly.

SARP>> if a bundle A is revealed preferred to bundle B directly/indirectly, we should never

observe B revealed preferred to bundle A directly or indirectly.

Main diff WARP & SARP: revealed preference is direct in WARP while it can be direct or
indirect in SARP.

Checking SARP
Diagram
Bundle 1 (first row) is revealed preferred to bundle 2 (first row) & bundle 2 (second row)
is revealed preferred to bundle 3 (second row).

Bundle 1 Bundle 2 Bundle 3

Bundle 1 is revealed preferred indirectly to bundle 3.

Bundle 1 Bundle 3
Bundle 3

Likewise bundle 3 (third row) is revealed preferred to Bundle 4 (third row) & bundle 4 (fourth
row) is revealed preferred to bundle 1 (fourth row).

Bundle 3 Bundle 4 Bundle 1

Thus, bundle 3 is revealed preferred to bundle 1 indirectly.

24
 Bundle 1 Bundle 3

..SARP is violated. For non-violation of SARP, bundle 3 must not be preferred to bundle 1 if
bundle 1 is already preferred to bundle 3 . Thus, consumer preferences are inconsistent.

Bundle 1 is revealed preferred to bundle 2 , which is in turn revealed preferred to bundle 3 .

Thus, bundle 1 is directly preferred to bundle 3 but bundle 3 is not preferred to bundle 1 directly
or indirectly. SARP not violated and consumer is consistent in his preferences.

Criticism of Revealed Preference Theory

i. Revealed preference theory ignores the possibility of indifference in the consumer’s
behavior. Likely that a consumer may face bundles equally desirable and may find it
difficult to choose between them.
ii. It assumes price as only factor influencing consumer’s behavior. In real world, consumers
are affected by a no of socio- economic considerations.
iii. Revealed preference is conditional: derivation of demand requires positive income
elasticity as a necessary condition for the negative price elasticity.
iv. Revealed preference theory is silent about Giffen Paradox in which income elasticity is
negative.
v. Revealed preference theory axiom that choice reveals preferences may not hold.
Consumer may choose a bundle in first time period and another in next period. e.g. a
consumer may buy more potatoes and less onions today due to their relative prices and
money income they own at the time of purchase. Next day, he may reverse the order of
choice due to the size of income in posse despite same relative prices.

25