BUSINESS ECONOMICS

Slides by CAO Thi Hong Vinh and LU Thi Thu Trang

1
Chapter 3:
CONSUMER BEHAVIOR
& DEMAND

2
— Required: Business Economics and
Managerial Decision Making, C.4-6
— Recommend: Economics for
Business and Management, C.2

3
 STRUCTURE
1. Consumer behavior
2. Demand analysis
3. Demand function estimation

4
 1. Consumer behavior
1.1. Maximize utility with budget
constraint
1.2. Characteristic approach
1.3. Behavioral approach

5
 1.1 Maximize utility

Utility?

Indifference curve?

6
 1.1 Maximize utility

Indifference curve: same level of

utility/satisfaction over dif. bundles of
goods
- Assumption: 2 substitutes, the more is better
- Characteristics?
- Slope?
7
 1.1 Maximize utility
Characteristics?
Assume: 2 substitutes, the more is better
- Parallel
- The farther from origin, the higher utility
- Slope downward
- Convex to the origin

8
Indifference Curve

9
Slope of Indifference Curve?
Changing along the curve
D -> B -> E: Same utility, different slope:
Steeper? Flatter?

10
Slope of Indifference Curve?
= Δ Qy/Δ Qx

11
Marginal Rate of Substitution (MRS)?

12
 MRS and IC?
Δ TUx = MUx * Δ Qx
Δ TUy = MUy* Δ Qy à Δ Qy/Δ Qx = - MUx/ MUy
=> Slope of Indifference Curve = - MRSxy

13
Marginal Rate of Substitution (MRS)?
- As we move down the indifference curve, the consumer is trading
more of one good in place of less of the other to keep her total
utility constant
- Measures how much of one good a consumer requires in order to
be compensated for a one-unit reduction in consumption of
another good
- MRS between 2 goods depends on their marginal utility
- Example: marginal utility of good X is twice the marginal utility
of good Y, then a person would need 2 units of good Y to
compensate for losing 1 unit of good X, and the marginal rate of
substitution equals 2.
MRSxy = MUx/MUy
14
 Maximizing Utility
— Highest IC possible

— Constraint?

2/16/25 15
Budget constraint:
M = Px Q x + Py Q y
M Px
Þ Qy = - Qx
Py Py

Slope?
Slope of budget line = -Px/Py

16
 Budget line?

2/16/25 17
 Optimal Consumption?
Slope of Indifference Curve =
Slope of Budget Line
- MRSxy = - Px/Py
-MUx/ MUy = - Px/Py
àMUx/ MUy = Px/Py
à Consumer Equilibrium:
The marginal rate of
substitution between any
two goods is equal to the
ratio of their prices

2/16/25 18
Income change?

19
Price change?

20
Price change?
= Substitution Effect + Income Effect

21
Criticism of indifference curve:
- How to set/change preferences?
- Static theory: not determine the path of change
- Imperfect Information to make decisions?
- Ordering of preferences is NOT just based on
utility, BUT moral preference/hierarchy of needs
- No interactions among individuals
- Only deal with private goods, consumed
instantly
22
1.2 CHARACTERISTIC APPROACH

Lancaster (1966):
Consumers make choices not only between distinct goods but
also similar goods but with dif. combinations of
characteristics
(ex: same 4 wheels of car, but many dif. body shapes, enginze
and other features)
- Consumers want goods for their inherent characteristics
- Characteristic is a property of goods that is relevant to
choice by consumers and generates utility

2/16/25 23
1.2 CHARACTERISTIC APPROACH

Lancaster (1966):

- The utility = function of the characteristics that the good

encompasses
- Consumer seeks to maximize utility, under the income
constraint

2/16/25 24
1.2 CHARACTERISTIC APPROACH
Lancaster (1966):
Assumptions:
— Each product will have more than one characteristic.
— Each product will have a mix of characteristics that will
vary by brand.
— Characteristics are measurable objectively.
— Products are divisible and do not have to be purchased
in whole units.
— Products (or brands) are substitutes for each other
despite containing differing combinations of
characteristics.
2/16/25 25
2/16/25 26
2/16/25 27
 1.3 BEHAVIOURAL APPROACH
— Behavioural economic theory
— Consumers are not perfect and fully rational decision maker
— Consumers utilize rules of thumb and decision routines to
overcome limited abilities and partial information available
— Consumer decisions = routine purchases or large or
infrequent purchases
- Routine purchases = previous experience + the position
the purchase has in their overall budget. (eggs/cheeses)
- Infrequent purchases = decision making process (car) à
behavioural economic theories attempt to model

2/16/25 28
 1.3 BEHAVIOURAL APPROACH
— Behavioural theories tend to be inductive in nature in
that they study and observe the decision making
processes used by consumers and deduce the decision
rules used.
— The consumer is viewed as following a process that
involves collecting information, processing information,
comparing and eliminating products and finally making
a choice.
à the rules and routines used in processing information
+ eliminating and selecting products for further
consideration.
2/16/25 29
 1.3 BEHAVIOURAL APPROACH
Decision-making cycle
- Recognition of the need to make a choice.
- Search for possible solutions to the problem.
- Evaluation of rival alternative courses of action.
- Choice by ranking alternatives in order of preference.
- Implementation of a chosen course of action.
- Hindsight by examining the outcome to see whether
outcome matched perception.
à To cope with these, decision makers develop ‘‘rules of
thumb’’ or ‘‘decision heuristics’’, which continue to be used
as long as they produce satisfactory results.
2/16/25 30
 1.3 BEHAVIOURAL APPROACH
Procedures for Making Choices
- Information-processing tasks can be viewed as
constructing a choice matrix with rival products on one
axis and relevant characteristics on the other
- The rules used by consumers to evaluate the
information are grouped under two headings:
+ compensatory rules that compare positive and negative
features
+ non-compensatory procedures that eliminate products
on a single criterion or absolute level of performance

2/16/25 31
2/16/25 32
2/16/25 33
 1.3 BEHAVIOURAL APPROACH

Consumers are:
- NOT perfectly rational + fully informed
individuals who make choices to maximize
utility
- boundedly rational and not fully informed
à make choices that satisfies their preferences

2/16/25 34
 2.Demand analysis
2.1 Market demand
2.2 Demand and Revenue
2.3 Elasticity

35
 2.1 Market demand
Function of:
Q x = f (Px , Py , A x , Y, T, O)
Q x : quantity demanded of good X
Px : price of good X
Py : price of good Y
A x : advertising expenditure
Y: real disposable income
T: consumer tastes
O: other factors
36
 2.1 Market demand
For simplicity:
Q x = f (Px )

i.e. Linear demand function: Q x = a + bPx

à Q = horizontal intercept: a
à P = vertical intercept: -a/b
à Slope ΔPx/ ΔQx = -1/b

37
 2.1. Market Demand
— Demand is the desire of a consumer to purchase a good
or service.
— The demand curve is a graphic representation of the
path along which the consumer would choose to
purchase quantities of the good or service at various
prices
— Market demand: the summation of individual demand
curves and shows the quantities of a product that would
be purchased by a group of consumers over a range of
possible prices.
2/16/25 38
 2.2 Demand and Revenue

The demand curve can also be used to

calculate total and marginal revenue.
i.e.
Q x = a + bPx
TR? MR?

39
 2.2 Demand and Revenue
i.e.
Q x = a + bPx
àa linear MR curve
à a MR curve: slope is twice that of the demand
curve
àTR is maximized where MR is 0

40
2.2 Demand and Revenue
Q x = a + bPx

41
 2.3 Elasticity
2.3.1. Own price elasticity of demand
2.3.2. Cross-price elasticity of demand
2.3.3. Income elasticity of demand
2.3.4. Advertising elasticity of demand

2/16/25 42
 2.3 Elasticity
2.3.1. Own price elasticity of demand
- Responsiveness of Demanded Quantity to changes in
price
- the ratio of the percentage change in quantity demanded
to the percentage change in the price that caused the
quantity change
DQ x / Q x
a. Formula: e=
DPx / Px
- Sign? -
- Magnitude? 1

43
Perfectly
Inelastic Inelastic Unit- Elastic
Demand Demand

Elastic Perfectly 44
Demand Elastic
 2.3 Elasticity
Q x = a + bPx
DQ x / Q x
e=
DPx / Px
MR = P (1– 1/e)?

e MR (0) Change in TR
(as P falls)
Inelastic - D

Elastic + I

45
46
 2.3 Elasticity
b. Factors affecting e :
Price effect = substitution effect + income effect
- Substitution effect: Closer substitute à more
or less elastic?
- Income effect: Large proportion of income à
more or less elastic?
- Time: Longer period à more or less elastic?

47
 2.3 Elasticity
c. Arc elasticity vs. Point elasticity:
Non-linear demand curve

48
 2.3 Elasticity
c. Arc elasticity vs. Point elasticity:
Arc elasticity
Q 1 + Q2
DQ/( )
e = 2
P1 + P2
DP / ( )
2

49
 2.3 Elasticity
c. Arc elasticity vs. Point elasticity:
Point elasticity

DQ x / Q x
e=
DPx / Px

50
 2.3 Elasticity
2.3.2. Cross-price elasticity
Responsiveness of demand for X to a
change in price of Y
Formula:
DQ x / Q x
eC =
DPy / Py

Substitute or complementary?

51
 2.3 Elasticity
2.3.3. Income elasticity
Responsiveness of demand for a product to a
change in income
Formula: DQ x / Q x
eI =
DI / I
Normal or inferior?
- Engel curve
- Factors affecting: initial income level, status of the
goods (necessities or luxuries), age of the goods…
52
2/16/25 53
Engel curves

54
 2.3 Elasticity
2.3.4. Advertising elasticity
Responsiveness of quantity demanded for
a change in Advertising expenditure
Formula:
DQ x / Q x
eA =
DA / A
Informative or persuasive?
Elastic or Inelastic?
55
 Exercise
— Exercise 1 and 2 of Review Questions

2/16/25 56
3. Demand function estimation
Different methods:
- Interviews and surveys (using
questionnaires)
- Consumer experiments
- Market studies
- Statistical analysis (using regression)

57
3. Demand function estimation
Statistical analysis (using regression)
3.1 Setting model:
Linear equation:
Q x = a + b1 Px + b 2 Py + b 3 A x + b 4 Y + b 5 X n
Log-linear equation:
logQ x = a + b1logPx + b 2 logPy + b 3 logA x
+ b 4 logY + b 5 logX n
58
3. Demand function estimation
Statistical analysis (using regression)
3.1 Setting model: Pitfalls
— Specification errors
- One or more important determinants of demand are
omitted from the model
- Wrong functional form was specified to estimate the
function
à Low value for R2 or Coefficients not having the
expected signs

59
3. Demand function estimation
Statistical analysis (using regression)
3.1 Setting model: Pitfalls
— Identification problems:
simultaneous change between
one variable included in the
model and one not included

60
3.2 Economic Verification
a. Check:
- Sign: if not the expected
- Magnitudes: outside the expected range
à proceed with caution

61
3.2 Statistical Verfication
b. Overall explanatory power of a regression model
R2 and adjusted R2

TSS- Total Sum of Squares

ESS- Explained Sum of Squares

RSS- Residual Sum of Squares

R2 = ESS/TSS = 1 - RSS/TSS à R2 = 1: perfect fit; R2 = 0: poor fit of OLS line

62
3. Demand function estimation
3.2 Statistical Verification
BUT: more variable -> higher R2 à using
adjusted R2

. Too low à misspecification of the model

(omission of imp. variables)
. Too high à multi-collinearity

63
3.2 Statistical Verification
c. F-test:
- testing the overall validity of the regression model
- dep. & a group of ind.
H0: no significant statistical rel. (null hypo)
Comparing: F-value estimated vs. benchmark value obtained from F distribution
statistical tables
Fb = function of degree of freedom of the denominator, the degree of freedom
for the numerator and the prob of being wrong.
The degree of freedom for the denominator = the number of independent
variables (excluding the constant term)
The degrees of freedom for the numerator = the number of observations
deducting the total number of independent variables including the constant
F > Fb à reject Ho: there is a significant relationship between the two variables.
64
2/16/25 65
3.2 Statistical Verification
d. t-test: dep. & an ind.
H0: no significant statistical rel. (null hypo)
comparing the t-value estimated when measuring the
regression relationship and the benchmark value obtained
from statistical tables.
te vs. tb
tb = Function of degree of freedom and the probability
of being wrong
te > tb à reject H0: significant relationship between
the two variables
à a general rule: tb is taken to be around 2 (prob 5%)
66
3.2 Econometric Verification
a. Multi-collinearity:
- Independent variables are correlated to each other (i.e. two independent
variables present the same information in a different way)
- High levels of multi-collinearity will have an effect on the estimated
coefficients.
- Detecting multi-collinearity = examining the R2 and t-statistics: the
model may achieve a high R2, but the t-statistics may indicate that a
number of variables, expected to be significant, are in fact insignificant.
- Solution = drop the less significant of the highly correlated variables or
to introduce time lags for some of the variables and to re-estimate the
model.

67
3.2 Econometric Verification
b. Auto-correlation:
Error terms: serially correlated
à Overestimating/underestimating the unexplained variation à accept
a good fit model but the result is dependent on autocorrelation
à Test = Durbin-Watson statistic:
H0: there is no auto-correlation (null hypothesis)
Compare DW to upper and lower limit derived from the statistical tables
DW > upper limit à H0 accepted
DW < lower limit à H0 rejected
(For the 5% level of significance with 25 data observations and 3
independent variables, the lower limit is 1.12 and the upper limit is
1.66)
- Solution: Adding other independent variables, transforming variables,
time lag of dependent variable, model specification 68
69