CHAPTER 3

Quantitative Demand Analysis

© 2017 by McGraw-Hill Education. All Rights Reserved. Authorized only for instructor use in the classroom. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
 Learning Objectives
1. Apply various elasticities of demand as a quantitative
tool to forecast changes in revenues, prices, and/or units
sold.
2. Illustrate the relationship between the elasticity of
demand and total revenues.
3. Discuss three factors that influence whether the demand
for a given product is relatively elastic or inelastic.
4. Explain the relationship between marginal revenue and
the own price elasticity of demand.
5. Show how to determine elasticities from linear and log-
linear demand functions.
6. Explain how regression analysis may be used to estimate
demand functions, and how to interpret and use the
output of a regression.
© 2017 by McGraw-Hill Education. All Rights Reserved. 2
 The Elasticity Concept
The Elasticity Concept
• Demand - Supply = Big picture, directional analysis.
• Magnitude of effects?
• e.g. How much will our sales change if rivals cut
their prices by 2 percent or a recession hits and
household incomes decline by 2.5 percent?
• Elasticities of demand as a quantitative forecasting
tool
• Elasticity
– A measure of the responsiveness of one variable to
changes in another variable; the percentage change in
one variable that arises due to a given percentage
change in another variable.

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-3

 The Elasticity Concept

The Elasticity Concept

• The elasticity between two variables, 𝐺 and 𝑆,
is mathematically expressed as:
%Δ𝐺
𝐸𝐺,𝑆 =
%Δ𝑆
• When a functional relationship exists, like 𝐺 =
𝑓 𝑆 , the elasticity is:
𝑑𝐺 𝑆
𝐸𝐺,𝑆 =
𝑑𝑆 𝐺
G = Qd (Quantity demanded) and S = P (Price)

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-4

 The Elasticity Concept

Measurement Aspects of Elasticity

• Important aspects of the elasticity:
– Sign of the relationship:
• Positive
• Negative
– Absolute value of elasticity magnitude relative to
unity:
• 𝐸𝐺,𝑆 > 1 𝐺 is highly responsive to changes in 𝑆.
• 𝐸𝐺,𝑆 < 1 𝐺 is slightly responsive to changes in 𝑆.

G = Qd (Quantity demanded) and S = P (Price)

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-5

 Own Price Elasticity of Demand

Own Price Elasticity of Demand

• Own price elasticity of demand
– Measures the responsiveness of a percentage change
in the quantity demanded of good X to a percentage
change in its price.
𝑑
%Δ𝑄𝑋
𝐸𝑄 𝑑
,𝑃𝑋 =
𝑋 %Δ𝑃𝑋
– Sign: negative by law of demand.
– Magnitude of absolute value relative to unity:
• 𝐸𝑄 𝑑
,𝑃𝑋 > 1: Elastic.
𝑋

• 𝐸𝑄 𝑑
,𝑃𝑋 < 1: Inelastic.
𝑋

• 𝐸𝑄 𝑑
,𝑃𝑋 = 1: Unitary elastic.
𝑋

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-6

 Own Price Elasticity of Demand

Linear Demand, Elasticity, and Revenue

Price Linear Inverse Demand: 𝑃 = 40 − 0.5𝑄
$40 Demand: 𝑄 = 80 − 2𝑄
• Revenue = $10
$30 20
40 = $600
$20 × 60 $800
$35 $30
$20
$10
• Elasticity: −2 × 60
20
40
−3
−1
= −0.333
$30
• Conclusion: Demand is inelastic.
elastic.
unitary elastic.
$25

$20 Observation: Elasticity

varies along a linear
$15 (inverse) demand curve

$10

$5
Demand

0 10 20 30 40 50 60 70 80 Quantity

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-7

 Own Price Elasticity of Demand

Total Revenue Test

• When demand is elastic:
– A price increase (decrease) leads to a decrease
(increase) in total revenue.
• When demand is inelastic:
– A price increase (decrease) leads to an increase
(decrease) in total revenue.
• When demand is unitary elastic:
– Total revenue is maximized.

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-8

 Own Price Elasticity of Demand

Perfectly Elastic and Inelastic Demand

Price
Demand

𝐸𝑄𝑋 𝑑 ,𝑃𝑋 = 0

Perfectly Demand
elastic 𝐸𝑄𝑋𝑑 ,𝑃𝑋 = −∞

Perfectly Inelastic Quantity

Aspirin vs. life saving drugs

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-9
 Own Price Elasticity of Demand

Factors Affecting the Own Price

Elasticity
• Three factors can impact the own price elasticity of
demand:
– Availability of consumption substitutes
• The more substitutes available for the good, the more elastic the demand for it
• Demand for broadly defined commodities tends to be more inelastic than the
demand for specific commodities (e.g., food vs. beef)

– Time/duration of purchase horizon

• more inelastic in the short term than in the long term (e.g., demand for a taxi of a
consumer who want to catch a flight)

– Expenditure share of consumers’ budgets

• Goods that comprise a relatively small share of consumers’ budgets tend
to be more inelastic than goods for which consumers spend a sizable
portion of their incomes (e.g., salt).

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-10

 Cross-Price Elasticity

Cross-Price Elasticity
• Cross-price elasticity
– Measures responsiveness of a percent change in
demand for good X due to a percent change in the
price of good Y.
𝑑
%Δ𝑄𝑋
𝐸𝑄 𝑑
,𝑃𝑌 =
𝑋 %Δ𝑃𝑌
– If 𝐸𝑄 𝑑 > 0, then 𝑋 and 𝑌 are substitutes.
𝑋 ,𝑃𝑌

– If 𝐸𝑄 𝑑
,𝑃𝑌 < 0, then 𝑋 and 𝑌 are complements.
𝑋

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-11

 Cross-Price Elasticity

Cross-Price Elasticity in Action

• Suppose it is estimated that the cross-price
elasticity of demand between clothing and
food is -0.18. If the price of food is projected
to increase by 10 percent, by how much will
demand for clothing change?
%∆𝑄𝐶𝑙𝑜𝑡ℎ𝑖𝑛𝑔 𝑑 𝑑
−0.18 = ⇒ %∆𝑄𝐶𝑙𝑜𝑡ℎ𝑖𝑛𝑔 = −1.8
10
– That is, demand for clothing is expected to decline
by 1.8 percent when the price of food increases
10 percent.

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-12

 Cross-Price Elasticity

Cross-Price Elasticity
• Cross-price elasticity is important for firms
selling multiple products.
– Price changes for one product impact demand for
other products.
• Assessing the overall change in revenue from
a price change for one good when a firm sells
two goods is:
∆𝑅 = 𝑅𝑋 1 + 𝐸𝑄 𝑑
,𝑃𝑋 + 𝑅𝑌 𝐸𝑄 𝑑
,𝑃𝑋 × %∆𝑃𝑋
𝑋 𝑌

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-13

 Cross-Price Elasticity

Cross-Price Elasticity in Action

• Suppose a restaurant earns $4,000 per week in
revenues from hamburger sales (X) and $2,000
per week from soda sales (Y).
• If the own price elasticity for burgers is 𝐸𝑄𝑋 ,𝑃𝑋 =
− 1.5 and the cross-price elasticity of demand
between sodas and hamburgers is 𝐸𝑄𝑌,𝑃𝑋 =
− 4.0, what would happen to the firm’s total
revenues if it reduced the price of hamburgers by
1 percent?
∆𝑅 = $4,000 1 − 1.5 + $2,000 −4.0 −1%
= $100
– That is, lowering the price of hamburgers 1 percent
increases total revenue by $100.

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-14

 Income Elasticity

Income Elasticity
• Income elasticity
– Measures responsiveness of a percent change in
demand for good X due to a percent change in
income.
𝑑
%Δ𝑄𝑋
𝐸𝑄 𝑑
,𝑀 =
𝑋 %Δ𝑀
– If 𝐸𝑄 𝑑 > 0, then 𝑋 is a normal good.
𝑋 ,𝑀

– If 𝐸𝑄 𝑑 < 0, then 𝑋 is an inferior good.

𝑋 ,𝑀

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-15

 Obtaining Elasticities From Demand Functions

Elasticities for Linear Demand

Functions
• From a linear demand function, we can easily
compute various elasticities.
• Given a linear demand function:
𝑑
𝑄𝑋 = 𝛼0 + 𝛼𝑋 𝑃𝑋 + 𝛼𝑌 𝑃𝑌 + 𝛼𝑀 𝑀 + 𝛼𝐻 𝑃𝐻
𝑃𝑋
– Own price elasticity: 𝛼𝑋 𝑑 .
𝑄𝑋
𝑃𝑌
– Cross price elasticity: 𝛼𝑌 𝑑 .
𝑄𝑋
𝑀
– Income elasticity: 𝛼𝑀 .
𝑄𝑋 𝑑

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-16

 Obtaining Elasticities From Demand Functions

Elasticities for Linear Demand

Functions In Action
The daily demand for Invigorated PED shoes is estimated to
be:
𝑄𝑋 𝑑 = 100 − 3𝑃𝑋 + 4𝑃𝑌 − 0.01𝑀 + 2𝑃𝐴𝑋
Suppose good X sells at $25 a pair, good Y sells at $35,
the company utilizes 50 units of advertising, and average
consumer income is $20,000. Calculate the own price,
cross-price and income elasticities of demand.
– 𝑄𝑋 𝑑 = 100 − 3 $25 + 4 $35 − 0.01 $20,000 +
2 50 = 65 units.
25
– Own price elasticity: −3( ) = −1.15.
65
35
– Cross-price elasticity: 4( ) = 2.15.
65
20,000
– Income elasticity: −0.01( ) = −3.08.
65
© 2017 by McGraw-Hill Education. All Rights Reserved. 3-17
 Obtaining Elasticities From Demand Functions

Elasticities for Nonlinear Demand

Functions
• One non-linear demand function is the log-
linear demand function:

ln 𝑄𝑋 𝑑
= 𝛽0 + 𝛽𝑋 ln 𝑃𝑋 + 𝛽𝑌 ln 𝑃𝑌 + 𝛽𝑀 ln 𝑀 + 𝛽𝐻 ln 𝐻

– Own price elasticity: 𝛽𝑋 .

– Cross price elasticity: 𝛽𝑌 .
– Income elasticity: 𝛽𝑀 .

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-18

 Obtaining Elasticities From Demand Functions

Elasticities for Nonlinear Demand

Functions In Action
An analyst for a major apparel company estimates that the
demand for its raincoats is given by
𝑙𝑛 𝑄𝑋 𝑑 = 10 − 1.2 ln 𝑃𝑋 + 3 ln 𝑅 − 2 ln 𝐴𝑌
where 𝑅 denotes the daily amount of rainfall and 𝐴𝑌 the
level of advertising on good Y. What would be the impact
on demand of a 10 percent increase in the daily amount
of rainfall?
%∆𝑄𝑋 𝑑 %∆𝑄𝑋 𝑑
𝐸𝑄 𝑑
,𝑅 = 𝛽𝑅 = 3. So, 𝐸𝑄 𝑑
,𝑅 = ⇒3= .
𝑋 𝑋 %∆𝑅 10

A 10 percent increase in rainfall will lead to a 30 percent

increase in the demand for raincoats.
© 2017 by McGraw-Hill Education. All Rights Reserved. 3-19
 Regression Analysis

Regression Analysis
• How does one obtain information on the
demand function?
– Published studies
– Hire consultant
– Statistical technique called regression analysis
using data on quantity, price, income and other
important variables.

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-20

 Regression Analysis

Regression Line and Least Squares

Regression
• True (or population) regression model
𝑌 = 𝑎 + 𝑏𝑋 + 𝑒
– 𝑎 unknown population intercept parameter.
– 𝑏 unknown population slope parameter.
– 𝑒 random error term with mean zero and standard deviation
𝜎.
• Least squares regression line
෠
𝑌 = 𝑎ො + 𝑏𝑋
– 𝑎ො least squares estimate of the unknown parameter 𝑎.
– 𝑏෠ least squares estimate of the unknown parameter 𝑏.
• The parameter estimates 𝑎ො and 𝑏, ෠ represent the values
of 𝑎 and 𝑏 that result in the smallest sum of squared
errors between a line and the actual data.
© 2017 by McGraw-Hill Education. All Rights Reserved. 3-21
 Regression Analysis

Excel and Least Squares Estimates

SUMMARY
OUTPUT
Estimated Demand:
Regression Statistics 𝑄 = 1631.47 − 2.60𝑃𝑅𝐼𝐶𝐸
Multiple R 0.87 𝑎ො = 1631.47
R Square 0.75
Adjusted R Square 0.72
𝑏෠ = −2.60
Standard Error 112.22
Observations 10.00

ANOVA
Df SS MS F Significance F
Regression 1 301470.89 301470.89 23.94 0.0012
Residual 8 100751.61 12593.95
Total 9 402222.50

Coefficients Standard Error t Stat P-value Lower 95% Upper 95%

Intercept 1631.47 243.97 6.69 0.0002 1068.87 2194.07
Price -2.60 0.53 -4.89 0.0012 -3.82 -1.37

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-22

 Regression Analysis

Evaluating Statistical Significance

• Standard error
– Measure of how much each estimated estimate
varies in regressions based on the same true
demand model using different data.
• 95 Percent Confidence interval rule of thumb
– 𝑎ො ± 2𝜎𝑎ො
– 𝑏෠ ± 2𝜎𝑏෠
• t-statistics rule of thumb
– When 𝑡 > 2, we are 95 percent confident the
true parameter is in the regression is not zero.

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-23

 Regression Analysis

Excel and Least Squares Estimates

SUMMARY
OUTPUT
෢ = 243.97
𝑠𝑒 (𝑎)
Regression Statistics
Multiple R 0.87
෢ = 0.53
𝑠𝑒(𝑏)
R Square 0.75 𝑡𝑎ො = 6.69 > 2, the intercept is different
Adjusted R Square 0.72 from zero.
Standard Error 112.22
𝑡𝑏෠ = −4.89 < 2, the intercept is different
Observations 10.00
from zero.
ANOVA
Df SS MS F Significance F
Regression 1 301470.89 301470.89 23.94 0.0012
Residual 8 100751.61 12593.95
Total 9 402222.50

Coefficients Standard Error t Stat P-value Lower 95% Upper 95%

Intercept 1631.47 243.97 6.69 0.0002 1068.87 2194.07
Price -2.60 0.53 -4.89 0.0012 -3.82 -1.37

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-24

 Regression Analysis

Evaluating the Overall Fit of the

Regression Line
• R-Square
– Also called the coefficient of determination.
– Fraction of the total variation in the dependent
variable that is explained by the regression.

2
𝐸𝑥𝑝𝑙𝑎𝑖𝑛𝑒𝑑 𝑉𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛 𝑆𝑆𝑅𝑒𝑔𝑟𝑒𝑠𝑠𝑖𝑜𝑛
𝑅 = =
𝑇𝑜𝑡𝑎𝑙 𝑉𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛 𝑆𝑆𝑇𝑜𝑡𝑎𝑙

– Ranges between 0 and 1.

• Values closer to 1 indicate “better” fit.
© 2017 by McGraw-Hill Education. All Rights Reserved. 3-25
 Regression Analysis

Evaluating the Overall Fit of the

Regression Line
• Adjusted R-Square
– A version of the R-square that penalize
researchers for having few degrees of freedom.
2 2
𝑛−1
𝑅 =1− 1−𝑅
𝑛−𝑘
– 𝑛 is total observations.
– 𝑘 is the number of estimated coefficients.
– 𝑛 − 𝑘 is the degrees of freedom for the
regression.

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-26

 Regression Analysis

Evaluating the Overall Fit of the

Regression Line
• The F- Statistic
• A measure of the total variation explained by
the regression relative to the total unexplained
variation.
– The greater the F-statistic, the better the overall
regression fit.
– Equivalently, the P-value is another measure of the
F-statistic.
• Lower P-values are associated with better overall
regression fit.
© 2017 by McGraw-Hill Education. All Rights Reserved. 3-27
 Regression Analysis

Excel and Least Squares Estimates

SUMMARY
OUTPUT

Regression Statistics
Multiple R 0.87
R Square 0.75
Adjusted R Square 0.72
Standard Error 112.22
Observations 10.00

ANOVA
Df SS MS F Significance F
Regression 1 301470.89 301470.89 23.94 0.0012
Residual 8 100751.61 12593.95
Total 9 402222.50

Coefficients Standard Error t Stat P-value Lower 95% Upper 95%

Intercept 1631.47 243.97 6.69 0.0002 1068.87 2194.07
Price -2.60 0.53 -4.89 0.0012 -3.82 -1.37

© 2017 by McGraw-Hill Education. All Rights Reserved. 3-28