please find an article from your favorite news source that covers the material covered

during that week (e.g., in the zoom audio lectures, power point presentations, the written transcripts etc.) (attached) and

explain in a paragraph or so how the article relates to that topic ( 2 paragraphs in total) .

Lecture on Optimization (Lecture 3-Week 3)

Todayâ€™s lecture materials can be found on blackboard within the folder titled, â€œLecture 3 (Optimization)Week 3â€ in â€œLecture Slidesâ€ under â€œCourse Documentsâ€.

Lecture Materials (This is what I would write on the left part of the white board during face-to-face

lecture):

1. Slides on optimization (Slides 1-17)

2. Break

3. Slides on optimization (Slides 18-24)

4. 1 in-class problem

(There are in-problems posted in the same folder (i.e., â€œLecture 3 (Optimization)-Week 3â€)

under the folder headed â€œIn-class problemsâ€ and the solution is also posted).

Todayâ€™s Lecture

We typically start each lecture with a theme. In broad sense, the term optimization refers to making the

best use out of a situation or resource. In our course, optimization refers to the process involved with

selecting the best decision from a set of alternatives. We can see that these two definitions are a lot

alike. They both refer to making the â€œbestâ€ decision and they both refer to selecting something or

making the best use out of a resource. These definitions guide the purpose of todayâ€™s lecture which is to

choose the â€œbestâ€ output from a set of alternatives. The word best is used in inverted quotes because

how we interpret what is the best depends upon our objective or what it is that we are trying to

achieve. We may be interested in selecting a strategic direction, minimizing cost, or maximizing profits.

These are all goals that involve the best output, as we mentioned during the first week of the course.

For the purposes of this course, we will focus on maximizing profit and maximizing revenue. We will also

focus on minimizing cost, but only within the context of maximizing profit. Regardless of what the

objective is, the manager would need to figure out how much output to produce in order to maximize

their earnings. The concept of the slope from algebra class from our high school classes will help us to

make this decision and gives way for our theme for todayâ€™s lecture. The theme in this lecture is that

slopes are useful to managers in helping them to make sound (i.e., well thought out) business decisions.

These ideas get introduced on slide 2 where find the purpose of optimization techniques and as

suggested above it provides us with a decision rule that enables us to choose from a set of alternatives.

A decision rule just guides us with how we are supposed to run our business given the economic

conditions at hand. An example would be Q=3 which is interpreted as: we need to produce 3 units-this is

the rule, to produce 3 units. The alternatives would be producing 4 units or 1 unit or 6 units etc. and Q

stands for Quantity of the control variable or as we see on slide 3, units of the control variable. You are

the manager and so you control or decide how much to produce. We use this variable to make

managerial type or business decisions. You can produce at any quantity you want, but the best thing you

can do for the operation, based upon economic theory, is to produce where your earnings objective is at

a maximum.

In slide 3, we also see some notation that will be useful throughout the semester. Before we dive into

this notation, we motivate this terminology to help us to remember these terms. When we produce

output, we hope to benefit from all the hard work and effort we put into making this output and this

benefit obtains from being able to sell your product in the market and earn income for your business.

Thus, we use the term benefit, denoted by B, as the sales or revenues earned from selling a certain

number of units denoted as Q and given in parentheses next to the B. The term B(Q), therefore, means

the benefits that I earn when I sell Q units. An example would look like the following: B(3) which means

the benefits that I earn when I produce and sell 3 units or B(6) which is interpreted as the benefits that I

earn when I produce and sell 6 units. We also assume that we are able to sell all the units we produce to

keep the analysis focused on production, output, and revenues or earnings (for instance, we do not then

consider inventory costs and how to incorporate these costs). C(Q) would then be, analogously, the

costs as they depend upon output produced. So, C(3) is the costs associated with producing 3 units and

C(6) is the costs associated with producing 6 units.

Slide 4 shows a picture of what the typical graph for benefits would look like for our class and during our

semester. Observe that it is an umbrella shape or upside down U-shape and can be thought of as â€“Q2

type of graph. Such a graph results from a linear demand curve as we also assume that total revenues

come from demand forecasts or from recognizing that if one is interested in revenues then one would

presumably analyze demand first.

Moreover and as motivated above, we would like to know, as managers, if we produce one more unit,

how much additional revenue can we earn? This is useful because assuming we have the resources by

producing one more unit if we can earn even 2 or 3 dollars, this is useful information. In slide 5, we

introduce the marginal benefit as the tool to help us here and observe that it is simply the slope of the

benefit function with respect to Q. It is interpreted as: if I produce one more unit of Q, how much more

revenue can I earn? Or equivalently, what is the benefit of producing one more unit of Q? Graphs of this

concept are given in slide 6 (in the form of the definition as rise/run) and 7 (in the form of the definition

as the tangent line slope at a single quantity value).

If benefit is B(Q) and cost is C(Q), then net benefit, i.e., the benefit after we adjust for cost or once we

take out cost is denoted as NB(Q)=B(Q)-C(Q). Once we subtract cost from benefit, we get net benefit

and since benefit is also revenue, net benefit is also profit. Managers would like to operate their

business where net benefit or profit is maximized. This information is provided in slide 8. In slide 9, we

see that this is where marginal profit is zero or where the slope of profit is a horizontal line. One point to

remember is that marginal profit is zero where total profit is maximized. Since NB(Q) = B(Q)-C(Q),

marginal NB(Q) = marginal B(Q) â€“ marginal C(Q) or MNB(Q) = MB(Q)-MC(Q). So, when MNB(Q)=0,

MB(Q)-MC(Q)=0 and so MB(Q)=MC(Q). The same above-mentioned concept works for revenue:

Marginal revenue is zero when total revenue is maximized. This concept is shown graphically in this slide

for profit.

Given the above discussion, we can arbitrarily select several values for Q and use these values to obtain

values for B(Q) and C(Q) and use the algebraic formulas to compute the respective slopes and try to use

these definitions to pick the Q that makes the slopes equal, i.e., MB(Q)=MC(Q). This could take a long

time as there are a potentially large number of Q values to go through. Fortunately, there are rules from

calculus that can greatly simplify this process. These rules are given on slide 11 through 13 and are slides

to remember from this slide deck.

Slide 11 is all about fixed costs. The marginal value for a fixed cost is zero. We will not deal with fixed

quantities in computing the revenues and so this slide is focused on costs. So, as an example, if C=3,

then no matter how much I produce my cost is still 3. So, if Q=1, C=3. If Q=2, C=3. By producing one

more unit, i.e., boosting production incrementally (moving from 1 unit of production to 2 units of

production) leads to no change in cost (cost was $3 initially when we produced 1 unit and it is still $3

when we produce 2 units). This means the slope of cost is: (3-3)/(2-1) = 0/1=0. When costs are fixed the

marginal contribution to cost through an incremental increase in production is zero. The implication for

businesses is that fixed costs should not factor into production decisions. Even if we shut down

operation, we still have to pay them.

Slide 12 is for the power rule. By power rule, we just mean, how do we compute the slope when we

have exponents. Some students like using the formula in the first line of the slide. To use this formula,

we need to identify what is a an what is n and then where we see DC/DQ, we would plug these numbers

into the formula to the right of the DC/DQ. So, if C = 2Q3 , then a =2 and n=3 and so DC/DQ = 3*2*Q3-1 =

6*Q2. A derivative is a rate of change and so if the initial cost equation has highest power of 3 then by

performing the operation that represents a first order rate of change (this just means we subtract 1

from the exponent), we should end up with the highest power of 2. So, if C=2Q, we first notice there is

no exponent. If there is no exponent, we can always explicitly assume there is a 1 and so C=2Q=2Q1.

Thus, a=2 and n=1. So, DC/DQ = 1*2*Q1-1 = 2*Q0 = 2*1 = 2. Finally, DC/DQ=MC =Marginal cost and this is

a definition. We always want to remember that slope=derivative=marginal, this is an important relation

to remember in this course.

Finally, in slide 13, we see the addition rule and this just says that if we want to find the slope of some

formula which adds together two terms, we need to find the slope of each term individually using slide

11 (for a fixed cost) or slide 12 (for a term with an exponent). Then, we need to add these slopes

together if we are told to add them and subtract these slopes, one from another, if we are told to do so.

So, in our example, C(Q)=4+3Q and so we first find the slope of each term individually and then because

the these two terms have a plus sign between them we would add the slopes together. So, since 4 does

not depend upon Q, we use slide 11 to determine that the slope of 4 with respect to Q is zero. Since 3Q

does have Q next to it, we use slide 12 to compute its relative marginal contribution to cost. So, 3*Q =

3*Q1 and so a=3 and n=1 which means MC = 1*3*Q1-1 = 3*Q0 = 3*1 = 3. Thus, the slope of C(Q) = 0 + 3 =

3.

Slides 14-17 help us to understand the concept of derivative a little better. In slide 14, we set up an

example to help us to see graphically that a derivative (these are the formulas we used on slide 11

through 13) is the same as a slope. We are just using the formulas on slide 11 through 13 to compute

the slope at each value of Q. For instance, if B=10Q-Q2, then MB = 10-2Q and the values for B are given

in this slide. When Q=1, B=10*1-12 = 10-1=9.

The next slide, slide 15, shows using this formula for marginal benefits that the slope at Q=4 for the

above equation for B using the slope ratio by subtracting a small amount from 4 and adding the same

small amount and computing the rise to run ratio at 4 but around this change. In either case, we get a

value for slope of 2.

If we computed the slope at each of points Q=1 to 5 in the same manner, i.e., using the formula for MB

above, we can compute the slope values as given in slide 16. In slide 17, we show these computations

using the graph from which we see that Q=5 corresponds to where MB=0 and this is where our revenues

are the largest at $25. We can also use equations to show this by setting MB=0 to get 10-2Q=0 and

solving for Q. MB=10-2Q=0 -> 10=2Q or 5=Q. Plugging 5 into the above equation for total benefits gives

us: B = 10*5-52 = 50-25 = 25.

Our in-class problem given in slide 18-24 is also given to you in the spreadsheet entitled â€œProfitMaximization Example-slide21-24.xlsxâ€ This is a good spreadsheet to look over and it is exactly what I

would write on the board if we were having a face to face class session. I encourage you to look over this

problem and to ask questions when and if any should come.

Optimization Techniques

SDSU – College of Business Administration

1

Why should we study Marginal

Analysis?

Optimization techniques provide a decision rule that will allow us to

choose the best decision from a set of alternatives.

Optimal managerial decisions involve comparing benefits to a decision

against the associated costs of that decision.

Marginal Analysis is the tool that allows us to compare incremental

benefits of a decision against the incremental costs of that decision

As managers, tools provided by Marginal Analysis will allow us to use

optimization techniques to make sound business decisions

âž¢ If incremental benefit of staying open exceeds the incremental cost of staying open,

then it makes sense to continue to stay open, otherwise, reverse. Similar logic

applies to the decision to take on a new project.

SDSU – College of Business Administration

2

Marginal Analysis: Introductory terms

The decisions of interest to managers often

involve revenues and costs, and so we will

generally refer to benefits as revenues and costs

as, well, costs.

We can define the benefits as B(Q) and the costs

then as C(Q), where Q is the number of units of

some control or decision variable (a variable that

we control or use to make a managerial decision)

A typical graph of B(Q) is shown on the next

slide. (This is the shape of all the graphs we will

consider in this class )

SDSU – College of Business Administration

3

Typical graph of Benefits

TR

Q

SDSU – College of Business Administration

4

Marginal Analysis: Slopes and

Economic terms

We might be interested in the additional

benefit (i.e. revenue) of adding one more

unit of some managerial control variable, Q,

on revenues.

The tool here is the marginal benefit and it

refers to the slope of the total revenue

function with respect to Q: Î”(TR)/Î”Q.

This is shown graphically on the next 2

slides

SDSU – College of Business Administration

5

Marginal Analysis: Slopes and economic

concepts

D TR

TR

DQ

Q

SDSU – College of Business Administration

6

Marginal Analysis: Slopes and

Economic Concepts

TR

DTR / D Q

Q

SDSU – College of Business Administration

7

Marginal Analysis: Slopes and

Economic Concepts

As managers, we want to maximize profits,

Ï€.

In this class, we will define profits to be

revenues â€“ costs. Symbolically, Ï€ = B(Q)C(Q).

Managers would like to operate their

business at the point where profits (or net

benefits) are maximized.

SDSU – College of Business Administration

8

Total profit

80000

Total profit

60000

Mp = 0

40000

20000

0

Slope = marginal profit

-20000

0

20

40

60

80

100

120

140

160

Output

SDSU – College of Business Administration

9

Marginal Analysis and Differential

Calculus

Given, the previous slides, we can plug in values

of Q to obtain values for total revenues and total

costs and hence profits and then we can develop a

table to see where profits are maximized.

This can be very time consuming, since we could

have a potentially large number of Q values to put

into the formulas to evaluate our revenue and cost

functions.

Fortunately, there are rules (i.e. formulas) which

help us determine the value of Q which enables us

to determine the maximum profits.

SDSU – College of Business Administration

10

Rules: Constants

C=a

dC / dQ = 0

Example:

C = 2

dC / dQ = 0

SDSU – College of Business Administration

11

Rules: Powers

C = aQn

dC / dQ = n a Q(n- 1)

Examples:

C = 2Q3

dC / dQ = (3)(2) Q(3 â€“ 1) = 6Q2

C = 2Q

dC / dQ = (1)(2) Q(1 â€“ 1) = 2

SDSU – College of Business Administration

12

Rules: Sums

C = f(Q) + g(Q)

dC / dQ = dC / dQ + dC / dQ

Example

C = 4 + 3Q

SDSU – College of Business Administration

dC / dQ = 3

13

A derivative is the same as a slope

30

B = 10Q â€“ Q2

24

25

21

20

16

10

9

0

0

0

1

2

SDSU – College of Business Administration

3

4

5

14

Calculate the slope around the point (4,24)

If Q = 3.9

If Q = 4.1

B = 10(3.9) â€“ (3.9)2 = 23.79

B = 10(4.1) â€“ (4.1)2 = 24.19

Slope = (24.19 â€“ 23.79) / (4.1 â€“ 3.9)

= 0.4 / 0.2 = 2

dB/dQ = 10 â€“ 2Q

At Q = 4 dB/dQ = 10 â€“ 2(4) = 2

SDSU – College of Business Administration

15

B = 10Q â€“ Q2

dB/dQ = 10 â€“ 2Q

At Q = 1

At Q = 2

At Q = 3

At Q = 4

At Q = 5

dB/dQ = 10 â€“ 2(1) = 8

dB/dQ = 10 â€“ 2(2) = 6

dB/dQ = 10 â€“ 2(3) = 4

dB/dQ = 10 â€“ 2(4) = 2

dB/dQ = 10 â€“ 2(5) = 0

SDSU – College of Business Administration

16

B = 10Q â€“ Q2

30

Slope = 4

20

Slope = 2

Slope = 6

10

Slope = 8

0

0

1

2

SDSU – College of Business Administration

3

4

5

17

Profit maximization example

Demand:

P = 1500 – 7Q

Q = 214.3 – 0.143P

Total Revenue:

TR = PQ = 1500Q – 7Q2

Total Cost:

TC = 225 + 38Q + Q2

SDSU – College of Business Administration

18

Total revenue, cost

100000

TR

80000

60000

40000

TC

20000

0

0

20

40

60

80

100

120

140

160

Output

SDSU – College of Business Administration

19

Profit maximization example

Demand:

P = 1500 – 7Q

Total Revenue:

TR = 1500Q – 7Q2

Total Cost:

TC = 225 + 38Q + Q2

Profit

p = [1500Q – 7Q2] â€“ [225 + 38Q + Q2 ]

= -225 + 1462Q – 8Q2

SDSU – College of Business Administration

20

Marginal profit approach

p = -225 + 1462Q – 8Q2

Mp = 1462 – 16Q

At Q = 10

p= -225 + 1462(10) â€“ 8(10)2 = $13,595

Mp = 1462 â€“ 16(10) = $1,302

SDSU – College of Business Administration

21

Marginal profit approach

Mp = 1462 – 16Q = 0

Q = 1462 / 16 = 91.375 =

91

TR = 1500(91) – 7(91)2 = $78,533

TC = 225 + 38(91) + (91)2 = $11,964

Profit = 78,533 – 11,964 = $66,569

SDSU – College of Business Administration

22

Marginal revenue, cost approach

Revenue:

TR = 1500Q – 7Q2

MR = 1500 – 14Q

Cost:

TC = 225 + 38Q + Q2

MC = 38 + 2Q

At Q = 10

MR = 1500 â€“ 14(10) = $1,360

MC = 38 + 2(10) = $58

SDSU – College of Business Administration

23

Marginal revenue, cost approach

Total Revenue:

Total Cost:

TR = 1500Q – 7Q2

TC = 225 + 38Q + Q2

MR = 1500 – 14Q

MC = 38 + 2Q

MR = MC

1500 – 14Q = 38 + 2Q

1462 = 16Q

Q = 91.375 = 91

SDSU – College of Business Administration

24

Bringing it all together!

Open the Excel Spreadsheet titled,

â€œSlide25-Problem-Week2â€

Look at the sheet titled â€œBasicsâ€

What is the first step to solving this

problem?

SDSU – College of Business Administration

25

Marginal Analysis Real World Example

Open the .pdf file,

â€œOptimizationApplication.pdfâ€

SDSU – College of Business Administration

26

Wrapping it all up!

Open the Excel Spreadsheet titled,

â€œOptimization-Inclass exercise-Set1.xlsâ€

I will do these two problems on board.

Open the Excel Spreadsheet titled,

â€œOptimization-Inclass exercise-Set2.xlsâ€

You can work with each other and I will

guide you.

SDSU – College of Business Administration

27

Appendix A: Product Rule

B = f(Q) g(Q)

dB/dQ = f(Q) Â· d[g(Q)]/dQ + d[f(Q)]/dQ Â· g(Q)

Example:

B = 3Q2 ( 3 – Q )

let f(Q) = 3Q2

g(Q) = 3-Q

dB / dQ = 3Q2 (- 1) + (3 – Q) (6Q)

= – 3Q2 + 18Q – 6Q2

= – 9Q2 + 18Q

SDSU – College of Business Administration

28

Appendix B: Extending to multiple

variables

Suppose revenue depends upon (1) price and

advertising or (2) sales of two products

â€œPartial derivativesâ€ describe the effect of a small

change on revenue

Process:

âž¢ Derive marginal functions for each decision

variable.

âž¢ Set each marginal function equal to zero.

âž¢ Solve for the decision variables.

SDSU – College of Business Administration

29

Revenue maximization

Revenue(x,y) = 200x + 100y – 10×2 – 20y2 + 20xy

ï‚¶Revenue / ï‚¶x = 200 – 20x + 20y = 0

ï‚¶Revenue / ï‚¶y = 100 – 40y + 20x = 0

200 – 20x + 20y = 0

100 + 20x – 40y = 0

300

– 20y = 0

300 – 20y = 0

y = 15

SDSU – College of Business Administration

200 – 20x + 20(15) = 0

x = 25

30

Revenue maximization

Revenue(x,y) = 200x + 100y – 10×2 – 20y2 + 20xy

x

23

24

25

26

27

y

15

15

15

15

15

Revenue

$3,210

$3,240

$3,250

$3,240

$3,210

SDSU – College of Business Administration

x

25

25

25

25

25

y

13

14

15

16

17

Revenue

$3,170

$3,230

$3,250

$3,230

$3,170

31

To find the equation of

the slope of the deman

VA-Intercept:

HA-Intercept:

Slope =

Elast. =

Region to the right of po

where we are unitary el

Region to the left of poi

where we are unitary el

If demand is elastic, an

price will lead to a redu

If demand is inelastic, a

will lead to an increase

If demand is unitary ela

revenues are maximize

These last three statem

is known as the total re

To determine the relationship between price, elasticity, and total revenue, we can also use the equation of the demand curve.

To find the equation of the demand curve, use the vertical axis intercept and the horizontal axis intercept to find

the slope of the demand curve

VA-Intercept:

HA-Intercept:

Slope =

Price =40 and Quantity=0

Price=0 and Quantity=80

-2

Q = -2P +b

Elast. =

-2* P/Q

Region to the right of point (Q>40)

where we are unitary elastic is inelastic

when Price=40, Q=0 -> b=80

when Price=0, Q=80 -> b=80

Equation is:

Region to the left of point (Q Q=40, P=20

If demand is unitary elastic (|E|=1), total

revenues are maximized

TR is Max when Q=40 and P=20

Or, when P/Q =1/2

These last three statements taken together

is known as the total revenue test

Q

P

0

10

20

30

40

50

60

70

P/Q

40

35

3.5

30

1.5

25 0.833333

20

0.5

15

0.3

10 0.166667

5 0.071429

TR

MR

0

350

600

750

800

750

600

350

30

20

10

0

-10

-20

-30

E

-7

-3

-1.66667

-1

-0.6

-0.33333

-0.14286

MR>0, elastic

MR |%Î”Q|>|%Î”P|->|E|>1

Unit Elastic -> |%Î”Q|=|%Î”P|=1

Inelastic -> |%Î”Q| 0 -> X and Y are substitutes

Â» Exy < 0 -> X and Y are complements

Cross-price elasticities play an important role in the pricing decisions

of firms that sell multiple products.

âž¢ E.g. selling a hamburger with a soda. These products are

complements-when a consumer purchases a burger, he or she will

typically purchase a soda with it.

SDSU – College of Business Administration

22

Other elasticities

Assume that firmâ€™s revenues are derived from sales of two products, X

and Y, then we can use the following formula to approximate changes

in revenues that obtain from changes in the price of product X:

Î”R = [Rx (1+EQx,Px) + RYEQY,Px ] x %Î”Px

Suppose that a restaurant earns $4,000 per week in revenues from

hamburger sales (Product X) and $2,000 per week from soda sales

(Product Y). Own price elasticity of demand for burgers is -1.5 and

cross price elasticity of demand between soda and burgers is -4.0.

What would happen to total revenues if it reduced the price of

hamburgers by 1%?

Homework set 3, Problem #6 will help you learn more about how

firms can take advantage of this idea and boost their revenues!

SDSU – College of Business Administration

23

Concept Check

A positive income elasticity tells us that that

good is:

âž¢a. a normal good

âž¢b. a substitute good

âž¢c. an inferior good

âž¢d. an inelastic good

The cross-price elasticity of demand

between bread and crackers is 4.

âž¢What can we conclude about bread and

crackers ?

SDSU – College of Business Administration

24

Elasticities from demand functions

Q = 6748 – 3P + 2.72Y + 1.68Po

Price

P = $15,000

Income

Y = $17,500

Related Price Po = $14,500

Q = 6748 – 3(15,000) + 2.72(17,500) + 1.68(14,500)

Q = 33,708

SDSU – College of Business Administration

25

Elasticities from demand functions

EP = -3 ( 15,000 / 33,708) = -1.335

EY = 2.72 ( 17,500 / 33,708 ) = 1.412

EPo = 1.68 ( 14,500 / 33,708 ) = 0.723

SDSU – College of Business Administration

26

TR and Price Elasticities

If you raise price, does TR rise?

Suppose demand is elastic, and raise price.

TR = Pâ€¢Q, so, %DTR = %DP+ %DQ

If elastic, P , but Q a lot

Hence TR FALLS !!!

Suppose demand is inelastic, and we decide

to raise price. What happens to TR?

SDSU – College of Business Administration

27

Relating TR to price elasticity

d TR

dP

= P + Q

dQ

dQ

Q dP

MR = P 1 +

P dQ

MR = P [ 1 + 1 / Ep ]

SDSU – College of Business Administration

28

Relating TR to price elasticity

MR = P [ 1 + 1 / Ep ]

E = -1

MR = 0

Implication: Revenues are maximized when E=-1.

|E| > 1

MR > 0

|E| < 1
MR < 0
SDSU - College of Business Administration
29
Relating TR to price elasticity
P = a â€“ bQ
TR = PQ = aQ - bQ2
MR = a - 2bQ
SDSU - College of Business Administration
30
Appendix A: Public Transportation
example
â€œHow prices and other factors affect travel behaviorâ€
http://www.vtpi.org/elasticities.pdf
Why would you want to use public transportation?
If using Metrolink, must have good reason (for e.g. not
many available substitutes-donâ€™t have alternative modes of
transportation)->inelastic demand

North County Transit District (NCTD) had been losing

ridership in 2010, and so it gambled on reducing fares to

increase ridership.

http://www.signonsandiego.com/news/2010/sep/15/nctdgamble-lower-fares-build-ridership/

SDSU – College of Business Administration

31

Appendix B: Constant elasticity demand

functions

Q = kPb

dQ / dP = bkPb-1

Ep =

dQ P

dP Q

= bkPb-1

P

kPb

E=b

SDSU – College of Business Administration

32

Constant elasticity demand functions

Q = P -2.5 Y 1.7 A 0.4

Ep = -2.5

EY = 1.7

EA = 0.4

Log-linear transformation

Log Q = -2.5(Log P) + 1.7(Log Y) + 0.4(Log A)

SDSU – College of Business Administration

33

Appendix C: Computing elasticity

without a demand function.

It turns out that if we are just given a range

of prices and quantities, we can do a

reasonable job computing elasticity using a

formula called Arc elasticity.

This can be useful for a manager who wants

to introduce a new product, but does not

have much data on sales or quantity

demanded.

SDSU – College of Business Administration

34

ARC Elasticities

%DQ

Q2 – Q1

(Q1 + Q2)/2

P2 – P1

(P1 + P2)/2

%DP

SDSU – College of Business Administration

35

Arc elasticity

20

19

12

14

SDSU – College of Business Administration

36

Arc elasticity

%DQ = 2 / 13 = 15.38%

%DP = -1 / 19.5 = -5.13%

20

E = -3.0

19

12

14

SDSU – College of Business Administration

37

Lecture on Theory of Elasticity of Demand (Lecture 8 (Elasticity of Demand)-Week 9&10)

Todayâ€™s lecture materials can be found on blackboard within the folder titled, â€œLecture 8 (Elasticity of

Demand)-Week 9&10â€ in â€œLecture Slidesâ€ under â€œCourse Documentsâ€.

Lecture Materials (This is what I would write on the left part of the white board during face-to-face

lecture):

1. Slides on Elasticity of Demand (Slides 1-26)

2. Break

3. Slides on Elasticity of Demand (Slides 27-30)

(covering the relationship between total revenue, elasticity of demand, and price)

Todayâ€™s lecture

All of the materials needed to follow along in this lecture are given to you under â€œLecture Materialsâ€.

With each lecture, we begin with a theme and the theme in todayâ€™s lecture is â€œby how muchâ€ the

concept of elasticity represents a sensitivity measure that allows you as the manager to build directly on

your toolkit from microeconomic theory of demand and in particular supply and demand. Therefore,

tools from supply and demand help us to determine trends in important economic variables like price

and income. When we use words like trends what we are saying is that we are able to qualitatively

define relationships between economic variables. For instance, prices go up and quantity demanded

comes down. Or, we know that for a normal good, an increase in income leads to an increase in quantity

demanded. In this lecture on elasticity, we take these ideas one step further and say okay we know that

a price of some good goes up, this means that the number of units of this good demanded by customers

goes down. This is by virtue of the Law of Demand. It would be useful for you as the manager to know

â€œby how muchâ€. So, in particular, if we know that prices for some good went up by 10%, it would be

useful for managers to know by how much (i.e., by how many units) did quantity demanded come

down? These questions are motivated on slide 2. This is the first theme for our lecture today, elasticity

tells us â€œby how muchâ€; it is a sensitivity measure. Please remember to check â€œclick to add notesâ€ for

more information.

In slide 3, we present an over of this lecture presentation. This lecture covers the price elasticity of

demand which is our first and basic measure of the quantitative effect of changes in price on the change

in quantity demanded. We begin by defining what is meant by this term and how we compute it. The

factors that determine price elasticity of demand are given on slides 16-20 and these slides will be for

your term project only and they will not be covered on your final exam project. So, in regards to the

term project, this presentation is helpful for question #7 and in specific question #7a and b which ask us

to describe the four factors (these are given in slide 16-20) that impact the elasticity of demand for our

firmâ€™s primary product and in part b we need to explain how these effects vary with the customer

segment defined in question 2c. Finally, we will wrap up our discussion in slides 27-30 and using the

spreadsheet titled â€œElast, MR, &TR.xlsxâ€.

The content of our lecture begins in slide 4 with what we call the â€œown price elasticity of demandâ€. It is

also called â€œprice elasticityâ€. This is measuring the effect on the number of units sold (we can also think

of this in terms of sales or revenue as this is defined as price x number of units sold) when there is a

change in price. It is also called â€œown price elasticity of demandâ€ because it measures the

responsiveness of the good for which we writing down the demand equation to changes in its own price.

This means that the effect of the price of a substitute or complementary good are not being measured,

but, rather, the effect of the price of the good for which we have written down the demand equation.

So, if we wrote down the demand equation for pizza, then this measure would assess the

responsiveness to changes in the price of pizza. Finally, elasticity is a unit less change (i.e., it is just a

number like 3 or 4 or 1.4) as it is defined in this slide as a percentage change. Again, it has the latin

terms ceteris paribus meaning all else equal.

ð¸ð‘ =

%Î”ð‘„

Î”ð‘„ ð‘ƒ

=

ð‘¥

%Î”ð‘ƒ

Î”ð‘ƒ ð‘„

Slope

Elasticity and slope are different. (See solution to study guide question 1 on page 48 for a good problem

that highlights this difference). A slope is a per unit change and elasticity is a percentage change. In slide

5, we define different degrees of sensitivity for the elasticity of demand. We define the elasticity of

demand to be elastic if |ð¸ð‘ | > 1, we use absolute value for interpretation but not computation. The

reason for this is because we know from the law of demand that an increase in price leads to a reduction

in quantity demanded and that this is the reason for the negative sign so we use the absolute value to

help us to focus on the interpretation through the magnitude. If we are elastic, then we are more price

sensitive. If |ð¸ð‘ | = 1 , we are unit elastic and this rarely happens in practice in business, but when

marginal revenues are zero and total revenues are maximized (as we saw on midterm #1), then we are

unit elastic, but only at this single points. Finally, if |ð¸ð‘ | < 1, then we are inelastic and in this case we
are less price sensitive. Goods that do not conform to laws of demand are called Veblan goods and these
correspond to certain luxury goods.
An example of an elastic demand is given on slide 6. The elasticity of pepsi is -2. So, immediately, we see
the negative sign and this is a result of the law of demand so we keep the sign for the purposes of
calculation but use absolute value for its interpretation in the following sense. |âˆ’2| = 2>1 and so since

the absolute value of the elasticity is greater than one, we have an elastic demand, this means we are

more price sensitive. What does this mean? If we were to increase the price of pepsi by 10%, then the

number of units sold of pepsi would be 20%. If we were to increase the price of pepsi by 1%, then the

number of units sold of pepsi would be 2%. This ratio is always 2 and so â€œmore price sensitiveâ€ means

that a percentage change in price would lead to a greater than exactly proportional percentage change

in quantity. If we are inelastic, the reverse would hold. A percentage change in price would lead to a

smaller than exactly proportional percentage change in quantity. Please try the question on slide 7 on

your own and the answer is d. This is an exactly proportional effect, meaning a percentage change in

price leads to an exactly proportional change in the percentage change in quantity. This is a unit elastic

demand curve.

A more complete list of definitions are given in slide 8. Examples of perfectly inelastic demand curves

correspond to products that are absolutely essential, perhaps even for living like medical equipment.

Certain types of equipment are needed for people to continue to live and so we are perfectly in sensitive

to changes in price. Perfectly elastic demand curves on the other side correspond to perfect substitutes

like for instance fruit. If you go to vons to get bananas or to ralphs and they do not have the same price

people should go to the cheaper price because a banana is banana it doesnâ€™t matter if Vons or Ralphs is

selling it. It may be different if its organic, but this is potentially a different product.

Sample elasticities are given in slide 9. Products that are considered more as necessities meaning we

need them for our every day life (e.g., eggs butter or chicken) would be more inelastic. Anything that is a

necessity would be more inelastic. Perishable goods like fruit tend to be more elastic. Ground beef is

close, but not exactly, unit elastic. We interpret that alcohol is a necessity from here also. Moving onâ€¦.

An explanation is needed here as we know that alcohol is not a necessity as is milk, butter or eggs. So,

instead of thinking about alcohol having an elasticity as below 1 as implying that it is a necessity, we

instead would think about it in terms of data and the pure mathematical definition for this good. What

an elasticity of demand that is less than one tells us is that the data on alcohol consumption indicate

that the consumption of alcohol is not very sensitive to price. This means that whether prices go up or

down does not seem to affect the purchase of alcohol. The reasoning behind why we tend to observe

this phenomena is a subject of research and this research is beyond the scope of the course content.

We now begin to discuss how to calculate elasticity. As we can infer from the equation representing the

definition of elasticity above, we need the slope and a price-quantity pair to compute elasticity. This

means that elasticity is calculated for each price-quantity pair given by the demand curve and so as we

move up or down the demand curve we calculate a different elasticity. This point is given in the first

bullet on slide 10. The second two bullet points we mentioned above when we defined elasticity and so I

refer you to that discussion for more information.

So, as is shown on slide 11, we would calculate elasticity by taking the product of the slope (which

comes from the demand equation) and the price-quantity pair. Either price or quantity must be given,

because if we have one we can use the demand equation to compute the other. Since the slope

represents a relative change of quantity with regard to price (holding all other factors constant) is it

simply interpreted as a per unit change in quantity relative to per unit change in price. If we have a

demand equation, then we would write this using the delta (the triangle looking symbol, we have been

using thus far) on slide 12. If we have are given a demand function (which includes all factors that you

think impact the sales for your product), then we represent the same quantity using a sea shell looking

symbol denoting a partial change. This symbol allows us to acknowledge that there are factors other

than price that impact our demand and we are holding them constant and this a partial change because

a total change would require changing all inputs and then seeing the affect on demand. It may not be as

informative from the standpoint of making a business decision because we do not know where the

source of the change in sales when we change three different inputs (as an example) at one time. So, we

isolate the impact of each underlying factor influencing quantity demanded using a sea shell looking

symbol given in slide 12 through the term on the extreme right most side of the statement of equalities.

So, in slide 13, if our demand equation is Q = 60 â€“ 10P and we want to find the price elasticity of demand

when price is 5 and quantity is 10; this means that P=5 and Q=10. We would begin with our definition of

elasticity which we repeat for convenience below:

%Î”ð‘„

Î”ð‘„

ð‘ƒ

ð¸ð‘ = %Î”ð‘ƒ = Î”ð‘ƒ ð‘¥ ð‘„

The first step would be to find the slope which is given in the demand equation. The slope is a relative

change meaning it tells a change in one variable relative to a change in another. So, here, since Q is in

the numerator and P is in the denominator, the slope tells us the change in Q relative to a change in P.

If I change price by one unit (i.e., one dollar or our measure of currency), how does quantity change

(i.e., the units demanded or sold or purchased). In a linear demand equation, all exponents are equal to

one and so the slope is just a constant equal to the number coming before the variable in the

denominator of our above expression, in this case P. Our slope is -10. To see this using more information

than just the demand equation and its slope. We can assume that P=$1 and compute Q. When P = $1, Q

= 60 â€“ 10*1=50. Then because we only want to know the relative change in Q due to the change in P

(this is the defn of slope), we only need to make P=$2, i.e., we only need to increment it by $1 to keep

the numbers simple in identifying the slope. So, when P=$2, Q = 60 â€“ 10*2=40. So, when we increase P

from $1 to $2, Q reduces from 50 units to 40 units; a $1 change in price leads to a 10 unit change in

quantity. The relative change in quantity due to a $1 change in price is 10 units. If you try other

numbers, you will still get 10, this is what it means to have a constant slope or rate or change. So, the

elasticity is going to be -10 *( P/Q). Here either P or Q has to be given. So, we are told P is 5, from the

demand equation Q = 60 â€“ 10*5 = 60-50=10, and so Q is 10 units. Even though we are told this, even if

we were not we could calculate it from the demand equation as shown in the prior sentence. So, our

elasticity is -10*(5/10)= -5. In absolute value this number is greater than 1 and so we are elastic at this

point on demand curve. We can calculate elasticities at each point of the demand curve in an analogous

manner and the numbers we would obtain are given in slide 14. Notice that as we move down the

demand curve (this means reducing price and increasing quantity) we move from being elastic to unit

elastic (only at a single point) to being inelastic. This will help us with addressing the second theme of

todayâ€™s lecture.

At higher values for price and necessarily (by law of demand) lower quantities, the ratio of P/Q is

relatively large and so with a constant slope this means the product of slope and P/Q would be a

relatively large number and this means its elastic. When prices are lower and necessarily quantities are

higher (by law of demand), the ratio of P/Q is relatively small and so with a constant slope this means

the product of slope and P/Q would be relatively a small number and this means its inelastic. Hopefully

between this explanation with words and the one given in the table on slide 14 this important

relationship between price, quantity, and elasticity is clearer. For managers, this relationship is

important for boosting sales which is where we will conclude this lecture shortly.

On slide 15, we add additional factors impacting the demand for our product to show how to compute

elasticity with a demand function. Its very similar as we illustrate below. So, as we have more variables

now to work through, we first compute Q using the demand equation and in this case what is given for

income and price. So, income=0 and price=5, plugging these into the demand function, we have: Q =

150-10*5+5*0 = 150-50+0=100. So, Q = 100, P=5, and the slope corresponding to the change in Q

relative to a change in P is again -10 (this is the coefficient before the P). Thus, the elasticity is: -10*

(5/100) =-10*(1/20)=-0.5. In absolute value, this is less than one and so we are inelastic. A one percent

change in price leads to a reduction in quantity demanded by half a percent.

Slide 16-20 are for your term project only and to help you, specifically, with question 7a and 7b. After

listing the four main factors affecting elasticity in slide 16, we begin with substitutes in slide 17 for which

we see that the better and more numerous the availability of substitutes the more elastic the demand

is. So, if we cannot really find substitutes for something like certain medical drugs or pharmaceutical

drugs needed to live, then we have an inelastic demand. Also, its hard to find a substitute for

transportation and so transportation has an inelastic demand; it is a broadly defined good. Analogously,

it is easy to find a substitute for a very specifically defined good like a Pontiac Catalina since it is easy to

find another product which delivers the same or similar value as a GM Pontiac Catalina. Whereas it is

not so easy to find another product which delivers the same or similar value as transportation since we

need some form of transportation to move from place to place and get to work for instance.

On slide 18, we see that when something costs more (as a proportion of our budget), the more elastic

the demand is, if you spend 15% of your income on something, you will be more sensitive to changes in

price. If you spend 1% of your income on something (like salt), you wont be as price sensitive. In slide 19,

we see that if you have more time to adjust to changes in price, then you are more elastic and as in the

longer term you have more time to seek out substitutes you tend to be more elastic in the longer term

versus the shorter term. So, gasoline is a good example here as when we first move to a city we may not

know it well enough and it may be hard to find a substitute but as we get acquainted with the city we

can find more suitable substitutes and this happens over time. The is the subject of the concept check in

slide 20 for which the answer is choice b. Demand tends to be more elastic in the longer term because

consumers have more time to seek out substitutes. Finally, the more durable or long lasting a product is,

the more elastic it is. So, cars are durable because used ones in many cases can be good substitutes for a

new one.

In slide 21 and 22, we discuss other elasticities. This includes elasticities related to income in slide 21

and related goods known as cross-price elasticities in slide 22. These are calculated the same way as

price elasticity, but now we do not use the absolute value because the sign plays an important role in

interpreting the result. For income elasticity, a positive sign means a normal good and a negative sign

means an inferior good and for cross-price elasticity, a positive sign means a substitute and a negative

sign means a complement. Other than this, these measures are computed in exactly the same way as

price elasticity of demand.

In slide 23, we show how cross-price elasticities play a role in pricing decisions for a firm with two goods.

So, lets say these goods are X and Y, where X =hamburgers and Y=sodas. The equation in bullet point

number 2 gives us a formula that allows us to use elasticities and cross-price elasticities to determine

revenues that can result from price changes. So, if ð‘…ð‘¥ =4,000 (this is the revenues from hamburger sales)

and ð‘…ð‘¦ =2,000 (this is revenue from soda sales) and EQx,Px = ð¸ð‘ = -1.5 and EQY,Px = EQdx,Py (using our

notation) = -4 (the negative sign means these numbers are complements which tend to be consumed

together. When someone gets a hamburger he/she tends to get soda with it. The interpretation of this

number is analogous to price elasticity of demand defined above). So, if we reduce the price of

hamburgers by 1%, then using the equation given in bullet point 2, we can find the effect on revenues as

shown below:

Î”R = [Rx (1+EQx,Px) + RYEQY,Px ] x %Î”Px

= [4,000(1+-1.5)+2,000(-4)]x-0.01

= [4,000(-0.5)-8,000]x-0.01

=[-2,000-8,000]x-0.01

=20+80

=100

So, when the business reduces the price of hamburgers and because when someone gets a hamburger

they tend to get a soda with it and the business earns more revenue from the soda sales (owing to cross

price elasticity) than the hamburger sales.

Slide 24 contains a concept check. If you have questions, please let me know, the answers are a and

substitutes.

Slide 25 is an example problem, please try this and let me know if you have questions. The answer is on

slide 26.

Finally, we conclude our lecture with our second and final theme. How do we use all these relationships

to boost our firm level revenue?

To help us address this theme, we turn to our spreadsheet entitled, â€œElast., MR, and TR.xlsxâ€ to help us

to visualize the graphs and equations needed to understand this theme as it relates to business.

We begin in slide 27 by addressing the question: when I increase price, what happens to total revenues?

So, via equations, in bullet point 2, we see how the percentage change in Q and percentage change in P

jointly affect the percentage change in revenue. This equation comes from the top of the sheet titled

â€œEquationsâ€ in the spreadsheet mentioned above. It is derived using the same percentage change

definition as elasticity except for revenue and after applying the chain rule in algebraic form and doing

more algebra to simplify the equation we end up with a relationship between total revenue, price, and

quantity. Our second equation which is underneath it is the definition of elasticity. Putting these two

equations together will allow us to answer our question posed at the beginning of this paragraph.

Two examples illustrating the relationship between price, elasticity, and revenue are given in the rest of

the sheet on the tab titled â€œEquationsâ€. So, in the first equation beginning underneath the box with the

two equations, we see that when we are elastic (E=-1.5) and we boost price (%Î”ð‘ƒ =6%), then %Î”Q =

Ex%Î”ð‘ƒ = -1.5 x 6% = -9% (from equation (2)). Then, percentage change in revenue will be the sum of the

percentage change in Q and P (this is from equation (1)) and this is -9%+6% = -3%. So, %Î”TR =-3%.

When we are elastic and we increase price we reduce total revenues. Think about traveling for leisure or

fun, when we are more sensitive to changes in price, if the airline company boosts prices, since we donâ€™t

need to travel, we will decide to postpone travel or to not travel and this reduces the revenues of airline

companies. This is shown the two graphs underneath of demand (with the ranges of elasticity) and

revenues with the big red arrows. When we are in the elastic region, we only look at the region on the

left where it says elastic. In this region, when we boost price (by law of demand this means reduce

quantity). The first red arrow and only red arrow in the demand equation plot is moving towards the top

of the price axis and towards the origin on the quantity axis in the elastic region. This means that

quantity is reducing and so we move to the origin of the revenue graph and this moves us away from the

top of the revenue graph and we reduce revenue. The reason we place these economic concepts on two

different graphs is because while the horizontal axis is quantity for both and this is measured in units,

the vertical axis for demand is price and this is measured in dollars per unit while the vertical axis for

revenues is dollars in sales and this measured in dollars. The units for the vertical axis variable is

different for demand and revenue and so they go on two different graphs.

When we are inelastic in the example below this so our elasticity , E=-0.5 and we boost price again

(%Î”ð‘ƒ=6%), then %Î”Q=-0.5×6=-3%. Change in revenues, %Î”TR=-3%+6%=3%. This means when we are

inelastic, and we boost price we boost revenues. If you have to travel to Dallas tomorrow morning for

work, even if the airline boost price since you have to go, you still go and the company still earns

revenue. We can see this graphically in the graph below the above described example. Now, we see the

red arrow on the demand plot on the left side since we are inelastic in this region, when we boost price

and reduce quantity we move in the same direction as in the example above except starting closer to

the right side of the graph rather than the left. In the revenue graph underneath we see that now we

start on the right side of the horizontal axis and reduce quantity and this moves us back towards the top

of the revenue graph thereby boosting revenues. When we are inelastic, we boost price then we

increase our revenues.

Finally, when we are unit elastic E=-1 and at this point and only at this point we have maximal revenues.

The marginal revenue values in these ranges is given in the tab called â€œGraphsâ€ which shows that when

we are elastic marginal revenues are positive when we are unit elastic marginal revenues are zero and

when we are inelastic marginal revenues are negative.

In slide 28, we use equations to show this relationship between marginal revenue, price, and elasticity

again using the chain rule from calculus. So, in slide 29, we see that when E=-1, MR=0 and when E0.

Finally in slide 30, we see that Demand and MR are, for linear equations, related via the slope and

intercept. The intercept is the same for demand and MR and the slope of MR is double the slope of

demand by a factor of 2. This is because we used a linear demand equation.

Purchase answer to see full

attachment