Monopolistic Monopolistic Competition & Competition &

OligopolyOligopoly

Monopolistic Monopolistic Competition & Competition &

OligopolyOligopoly

PGP-ME SESSION 10PGP-ME SESSION 10July 26, 2010July 26, 2010

PROF. SAMAR K. DATTAPROF. SAMAR K. DATTA

Checklist for students• Characteristics of Monopolistic Competition vis-à-vis

Monopoly

• Supernormal profit/loss in short run, but only normal profit in long-run equilibrium

• Still different from Perfectly Competitive Equilibrium: P>MC, presence of Excess Capacity & Deadweight Loss

• Efficiency Considerations under Monopolistic Competition• Conditions for Oligopoly• Role of Strategic Interdependence• Profit Maximization in Oligopoly Settings

– Cournot Model– Stackelberg Model – Bertrand Model

• Competition versus Collusion & Prisoners’ Dilemma2ME Slot 2

Characteristics of Monopolistic Competition

– Many firms

– Free entry and exit

– Differentiated but highly substitutable products (hence conjectural variation is close to zero, if not exactly zero)

• Amount of monopoly power depends on the degree of differentiation)

• The greater the consumer preference towards differentiated product, the higher the price

3ME Slot 2

A Monopolistically CompetitiveFirm in the Short and Long Run

Quantity

$/Q

Quantity

$/QMC

AC

MC

AC

DSR

MRSR

DLR

MRLR

QSR

PSR

QLR

PLR

Short Run Long Run

4ME Slot 2

• Short-run– Downward sloping demand--

differentiated product, demand is relatively elastic--good substitutes

– MC=MR < P => Firm makes economic profits

• Long-run– Profits will attract new firms to the

industry (no barriers to entry)– No economic profit (P = AC), but P >

MC => some monopoly power

A Monopolistically CompetitiveFirm in the Short and Long Run

5ME Slot 2

Deadweight lossMC AC

Comparison of Monopolistically CompetitiveEquilibrium and Perfectly Competitive

Equilibrium

$/Q

Quantity

$/Q

D = MR

QC

PC

MC AC

DLR

MRLR

QMC

P

Quantity

Perfect Competitive Firm Monopolistically Competitive Firm

6ME Slot 2

Monopolistic Competition and Economic Efficiency

– The monopoly power (differentiation) yields a higher price than perfect competition.

– Although there are no economic profits in the long run, the firm is still not producing at minimum AC and excess capacity exists.

– Variety in closely substitutable products, monopolistically competition offers to consumers to choose from, is also a dynamic efficiency gain to the society.

7ME Slot 2

Monopolistic Competition

The Good (To Consumers)– Product Variety

The Bad (To Society)– P > MC– Excess capacity

• Unexploited economies of scale

The Ugly (To Managers)– P = ATC > minimum of

average costs.• Zero Profits (in the

long run)!

ME Slot 2

Characteristics of Oligopoly

• Small number of firms• Product differentiation may or may not

exist• Barriers to entry

– Natural• Scale economies

• Patents

• Technology

• Name recognition– Strategic action

• Flooding the market

• Controlling an essential input

9ME Slot 2

Oligopoly Environment• Relatively few firms, usually less than

10.– Duopoly - two firms– Triopoly - three firms

• The products firms offer can be either differentiated or homogeneous.

• Firms’ decisions impact one another.• Many different strategic variables are

modeled:– No single oligopoly model.

10ME Slot 2

Equilibrium in a Oligopolistic Market

• In oligopoly the producers must consider the response of competitors when choosing output and price.

• Defining Cournot-Nash Equilibrium– Equilibrium: Firms do the best they can

and have no incentive to change their output or price

– Cournot-Nash Equilibrium: Each firm is doing the best it can, given what its competitors are doing

11ME Slot 2

First Mover Advantage--The Stackelberg Model

• Assumptions– One firm can set output first– Firm 1 sets output first (i.e., acts as leader)

and Firm 2 then makes an output decision (i.e., acts as follower)

– So, there are 2x2 contingencies – each firm behaving like either a leader or a follower, Cournot-Nash being only a special case (follower-follower) of Stackelberg model

12ME Slot 2

An example• Demand curve: P = 500 – 0.5X, where X = X1

+ X2• AC1 = MC1 =100; AC2 = MC2 = 150• Can the second fellow survive if there is price

competition?• Can he survive if there is quantity

competition as in Cournot-Nash model?• TR1 = (500 – 0.5X).X1 = 500X1 – 0.5(X1 +

X2)X1 = 500 -0.5X12 – 0.5X1.X2 => MR1 = dTR1/dX1 = 500 – 2.(1/2).X1 – 0.5X2 – 0.5X1.(dX2/dX1) = 500 – X1 -0.5X2, assuming conjectural variation =dX2/dX1 =0

13ME Slot 2

Π-maximization by firms I & II

• MR1 = MC1 => 500 – X1 – 0.5X2 = 100 => X1 = 400 -0.5X2 ----(a) giving the reaction function of firm 1, i.e., profit-maximizing value of X1 as function of X2.

• (a) => If X1 = 0, then X2 = 800; If X2 =0, then X1 = 400

• Similarly, MR2 = MC2 gives 500 –X2 -0.5X1 = 150 => X2 = 350 -0.5X1 ---(b), giving the reaction function of firm 2

• (b) => If X1 =0, then X2 = 350; If X2 =0, then X1 = 700

14ME Slot 2

Situation 1: Cournot-Nash Equilibrium

• Both firms are dormant (followers):– (a) & (b) => X1=300; X2=200; X=500; P=250;

Π1=45,000; Π2=20,000; Π=65,000; Thus firm-2 in profit, though relatively inefficient.

700400

200

350

800

X1

X2

(a)

(b)

533300

(c)

450125

Collusive solution: Max Π=(500-0.5x)x – 100x =400x – 0.5x2

=> dΠ/dx = 400 –x =0 =>x=400. Why is this solution lying on Firm 1’s reaction curve (a) where X1=400 and X2=0?

Attempting a competitive solution with MC1=AC1=100 & MC2=AC2=150: Firm-2 thrown out in equilibrium and firm-1 will charge price P=150-ε, which is marginally less than 150 to prevent firm-2 from surfacing. Thus, P=500 – 0.5x1 =>150 = 500 – 0.5x1 => x1 =700. Why is it lying on firm-2’s reaction curve (b)? Is it more meaningful to call it a monopoly solution or a contestable market solution?

15ME Slot 2

Situation 3 (Stackelberg): Firm II active (leader), Firm I dormant

(follower)

• Now MR2 = 500 –X2 -0.5(X1 + (dX1/dX2).X2) = 500 -3/4X2 – ½ X1

• Hence MR2=MC2 => 500 -3/4X2 – ½ X1 = 150 => X2 = 1400/3 – 2/3X1 ------(d), new reaction function of firm II

• Solving (a) & (d) gives– X1=250; X2=300; X=550– P=225– Π1=31,250; Π2=22,500; Π=53,750Thus, the relatively inefficient firm II increases market

share & profit (though still less than that of firm I in absolute terms).

Lesson: Can we allow our generally inefficient PSUs tocompete?

16ME Slot 2

Situation 4 (Stackelberg): Both firms active

(leaders)

• Here we can check that– X1=400; X2=200; X=600– P=200– Π1=40,000; Π2=10,000; Π=50,000

• Thus, even larger total output at even lower price

17ME Slot 2

All 4 situations put together in game theory format

Player 1 Player 2Dormant Active

Dormant 45,000; 20,000 31,250; 22,500

Active 50,625; 7,812 40,000; 10,000

To become active thus becomes the dominant strategy for both players

18ME Slot 2

Price Competition (homogenous good) – The

Bertrand Model• If two duopolists producing a homogenous good

compete by simultaneously choosing price, the good being homogenous, consumers will buy from the lowest price seller– The lower priced firm will supply the entire market

and the higher priced firm will sell nothing• Competitive price cutting by the firms will lead to the

perfectly competitive outcome• If both firms charge the same price, consumers will be

indifferent between firms and each firm will supply half the market.

• When firms produce a homogenous good, it is more natural to compete by setting quantities rather than prices (bringing us back to the Cournot model)

• 19ME Slot 2

• Assume:

16$ 6$ :Collusion

12$ 4$ :mEquilibriuNash

212 :demand s2' Firm

212 :demand s1' Firm

0$ and 20$

12

21

P

P

PPQ

PPQ

VCFC

Competition Versus Collusion: The Prisoners’

Dilemma

20ME Slot 2

Price Competition – Differentiated

Products

• Determining Prices and Output– Firm 1: If P2 is fixed:

12

21

2111

413

413

0412

'

PP

PP

PPP

curve reaction s2' Firm

curve reaction s1' Firm

price maximizing profit s1 Firm

21ME Slot 2

Firm 1’s Reaction Curve

Bertrand Model – Heterogeneous Good

Case

P1

P2

Firm 2’s Reaction Curve

$4

$4

Nash Equilibrium

$6

$6

Collusive Equilibrium

22ME Slot 2

• Possible Pricing Outcomes:

–

–

$16 $6, charge both If

Competition Versus Collusion:

The Prisoners’ Dilemma

4$204)6)(2(12)6(

20

20$206)4)(2(12)4(

20

4$6$

111

222

21

then

and If

QP

QP

PP

23ME Slot 2

Payoff Matrix for Pricing Game

Firm 2

Firm 1

Charge $4 Charge $6

Charge $4

Charge $6

$12, $12 $20, $4

$16, $16$4, $20

24ME Slot 2

• These two firms are playing a non-cooperative game.– Each firm independently does the

best it can taking its competitor into account.

• An example in game theory, called the Prisoners’ Dilemma, illustrates the problem oligopolistic firms face.

Competition Versus Collusion:The Prisoners’ Dilemma

25ME Slot 2

• Scenario– Two prisoners have been accused

of collaborating in a crime.– They are in separate jail cells and

cannot communicate.– Each has been asked to confess

to the crime.

The Prisoners’ Dilemma

26ME Slot 2

-5, -5 -1, -10

-2, -2-10, -1

Payoff Matrix for Prisoners’ Dilemma

Prisoner A

Confess Don’t confess

Confess

Don’tconfess

Prisoner B

27ME Slot 2

Implications of the Prisoners’

Dilemma for Oligipolistic Pricing

• In some oligopoly markets, pricing behavior in time can create a predictable pricing environment and implied collusion may occur.

• In other oligopoly markets, the firms are very aggressive and collusion is not possible.

• Firms are reluctant to change price because of the likely response of their competitors.

• In this case prices tend to be relatively rigid, leading to a kinked-demand curve model

28ME Slot 2

OLIGOPOLY-IIOLIGOPOLY-IIOLIGOPOLY-IIOLIGOPOLY-IIME SESSION 11ME SESSION 11

99THTH August, 2010 August, 2010

PROF. SAMAR K. DATTAPROF. SAMAR K. DATTA

Overview• Revisiting Cournot-Nash model with

trivial textbook example• Implications of cooperation (i.e.,

collusive agreement) & non-cooperation (i.e., competition)

• Kinked demand curve model• Price signaling & price leadership• Dominant firm model• Cartels

30ME Slot 2

Text book example of Cournot

Equilibrium

– Market demand is P = 30 - Q where Q = Q1 + Q2

– Assumption of MC1 = MC2 = 0 has a limitation not merely because it is trivially fixed at zero, but also because MCs are equal

– Firm 1’s Reaction Curve:

111 )30( Revenue, Total QQPQR

122

11

1211

30

)(30

QQQQ

QQQQ

31ME Slot 2

Cournot Equilibrium

12

21

11

21111

2115

2115

0

230

QQ

QQ

MCMR

QQQRMR

Curve Reaction s2' Firm

Curve Reaction s1' Firm

1030

20

10)2115(2115

21

2111

1

QP

QQQ

QQQQ

QQ 2:mEquilibriu Cournot

32ME Slot 2

Textbook Duopoly Example

Q1

Q2

Firm 2’sReaction Curve

30

15

Firm 1’sReaction Curve

15

30

10

10

Cournot Equilibrium

The demand curve is P = 30 - Q andboth firms have 0 marginal cost.

33ME Slot 2

First Mover Advantage--The Stackelberg Model

• Assumptions– One firm can set output first– MC = 0 as a simplifying assumption– Market demand is P = 30 - Q where Q

= total output– Firm 1 sets output first and Firm 2

then makes an output decision

34ME Slot 2

• Firm 1– Must consider the reaction of Firm 2

• Firm 2– Takes Firm 1’s output as fixed and

therefore determines output with the Cournot reaction curve: Q2 = 15 - 1/2Q1

First Mover Advantage--The Stackelberg Model

35ME Slot 2

• Firm 1

– Choose Q1 so that:

122

1111 30

0

Q - Q - QQ PQ R

MC, MC MR

0 MR therefore

First Mover Advantage--The Stackelberg Model

36ME Slot 2

• Substituting Firm 2’s Reaction Curve for Q2:

5.7 and 15:0

15

21

1111

QQMR

QQRMR

211

112

111

2115

)2115(30

QQ

QQQQR

First Mover Advantage--The Stackelberg Model

ConclusionFirm 1’s output is twice as large as firm 2’sFirm 1’s profit is twice as large as firm 2’s

37ME Slot 2

Why first-mover advantage?

• First-mover’s output would be large.• If second-mover too produces a large

output, then prices are driven down, causing loss to both.

• If however, the second-mover produces a small output, then both firms make profits, but the first-mover makes larger profits corresponding to his larger output.

• Since a rational second-mover values profits higher than revenge, he will set a small output, and both firms profit.

38ME Slot 2

Firm 1’sReaction Curve

Firm 2’sReaction Curve

Implications of the Duopoly Example in the

textQ1

Q2

30

30

10

10

Cournot Equilibrium15

15

Competitive Equilibrium (P = MC; Profit = 0)

CollusionCurve

7.5

7.5

Collusive Equilibrium

For the firm, collusion is the bestoutcome, followed by the Cournot

equilibrium and then the competitive equilibrium

39ME Slot 2

Implications of the Prisoners’

Dilemma for Oligipolistic Pricing

• In some oligopoly markets, pricing behavior in time can create a predictable pricing environment and implied collusion may occur.

• In other oligopoly markets, the firms are very aggressive and collusion is not possible.

• Firms are reluctant to change price because of the likely response of their competitors.

• In this case prices tend to be relatively rigid, leading to a kinked-demand curve model

40ME Slot 2

The Kinked Demand Curve Model

$/Q

D

P*

Q*

MC

MC’

So long as marginal cost is in the vertical region of the marginal

revenue curve, price and output will remain constant.

MR

Quantity41ME Slot 2

Price Signaling and Price Leadership

• Price Signaling

– Implicit collusion in which a firm announces a price increase in the hope that other firms will follow suit

• Price Leadership

– Pattern of pricing in which one firm regularly announces price changes that other firms then match

42ME Slot 2

The Dominant Firm Model

• In some oligopolistic markets, one large firm has a major share of total sales, and a group of smaller firms supplies the remainder of the market.

• The large firm might then act as the dominant firm, setting a price that maximized its own profits.

43ME Slot 2

Price Setting by a Dominant Firm

Price

Quantity

D

DD

QD

P*

At this price, fringe firmssell QF, so that total

sales are QT.

P1

QF QT

P2

MCD

MRD

SF The dominant firm’s demandcurve is the difference between

market demand (D) and the supplyof the fringe firms (SF).

44ME Slot 2

Cartels• Characteristics

– Explicit agreements to set output and price

– May not include all firms

– Most often international

– Conditions for success

• Competitive alternative sufficiently deters cheating

• Potential of monopoly power--inelastic demand

• Either the cartel must control nearly all of the world’s supply or the supply of non-cartel producers must not be price elastic 45ME Slot 2

The OPEC Oil CartelPrice

Quantity

MROPEC

DOPEC

TD SC

MCOPEC

TD is the total world demandcurve for oil, and SC is the

competitive supply. OPEC’s demand is the difference

between the two.

QOPEC

P*

OPEC’s profits maximizingquantity is found at the

intersection of its MR andMC curves. At this quantity

OPEC charges price P*.

46ME Slot 2

Cartels• About OPEC

– Very low MC– TD is inelastic– Non-OPEC supply is inelastic

– DOPEC is relatively inelastic

47ME Slot 2

The CIPEC Copper CartelPrice

Quantity

MRCIPEC

TD

DCIPEC

SC

MCCIPEC

QCIPEC

P*PC

QC QT

•TD and SC are relatively elastic•DCIPEC is elastic•CIPEC has little monopoly power•P* is closer to PC

48ME Slot 2

PRICING PRICING STRATEGIES FOR STRATEGIES FOR

FIRMS WITH FIRMS WITH MARKET POWER-IMARKET POWER-I

PRICING PRICING STRATEGIES FOR STRATEGIES FOR

FIRMS WITH FIRMS WITH MARKET POWER-IMARKET POWER-I

ME, SESSION 12ME, SESSION 12

10th August, 201010th August, 2010

PROF. SAMAR K. DATTAPROF. SAMAR K. DATTA

ChecklistI. Basic Pricing Strategies

– Markup Pricing

II. Extracting Consumer Surplus thro’ Charging Multiple Prices rather than a Single Price– Price Discrimination

• Price discrimination in a global market• Utility of discriminating monopolist

– Two-Part Pricing– Block Pricing– Commodity Bundling

III. Pricing for Special Cost and Demand Structures– Inter-temporal Price Discrimination: Peak-Load

Pricing

50ME Slot 2

Standard Pricing and Profits for Firms with

Market PowerPrice

Quantity

P = 10 - 2Q

10

8

6

4

2

1 2 3 4 5

MC

MR = 10 - 4Q

Profits from standard pricing= $8

51ME Slot 2

An Algebraic Example

• P = 10 - 2Q• C(Q) = 2Q• If the firm must charge a single price to

all consumers, the profit-maximizing price is obtained by setting MR = MC.

• 10 - 4Q = 2, so Q* = 2.• P* = 10 - 2(2) = 6.• Profits = (6)(2) - 2(2) = $8.

52ME Slot 2

Pricing with market power: the underlying idea

• Pricing with market power enables the producer to capture some of the consumer surplus.

• But it requires the individual producer to know much more about the characteristics of demand as well as the capability to manage production and the supply chain.

53ME Slot 2

A Simple Markup Rule• Suppose the elasticity of demand

for the firm’s product is EF.• Since MR = P[1 + EF]/ EF.• Setting MR = MC and simplifying

yields this simple pricing formula:P = [EF/(1+ EF)] MC.

• The optimal price is a simple markup over relevant costs!– More elastic the demand, lower markup.– Less elastic the demand, higher markup.

54ME Slot 2

An Example• Elasticity of demand for Kodak film is -

2.

• P = [EF/(1+ EF)] MC

• P = [-2/(1 - 2)] MC• P = 2 MC• Price is twice marginal cost.• Fifty percent of Kodak’s price is margin

above manufacturing costs.

55ME Slot 2

Markup Rule for Cournot Oligopoly

• Homogeneous product Cournot oligopoly.• N = total number of firms in the industry.

• Market elasticity of demand EM .

• Elasticity of individual firm’s demand is given by EF = N x EM.

• Since P = [EF/(1+ EF)] MC,

• Then, P = [NEM/(1+ NEM)] MC.

• The greater the number of firms, the lower the profit-maximizing markup factor.

56ME Slot 2

An Example• Homogeneous product Cournot

industry, 3 firms.• MC = $10.• Elasticity of market demand = - ½.• Determine the profit-maximizing price?

• EF = N EM = 3 (-1/2) = -1.5.

• P = [EF/(1+ EF)] MC.

• P = [-1.5/(1- 1.5] $10.• P = 3 $10 = $30.

57ME Slot 2

Extracting Consumer Surplus: Moving From Single Price Markets

• Most models examined to this point involve a “single” equilibrium price.

• In reality, there are many different prices being charged in the market.

• Price discrimination is the practice of charging different prices to consumer for the same good to achieve higher prices.

58ME Slot 2

Capturing Consumer Surplus

Quantity

$/Q

D

MR

Pmax

MC If price is raised above P*, the firm will lose

sales and reduce profit.

PC

PC is the pricethat would exist in

a perfectly competitivemarket.

A

P*

Q*

P1

Within [0-Q*], consumers willingto pay more than P*-- indicated by consumer surplus (area A).

B

P2

To sell above Q*, price willhave to fall to create a consumer surplus (see B).

59ME Slot 2

Price Discrimination• Price discrimination is the charging of

different prices to different consumers for similar goods.

– First degree price discrimination• Charge a separate price to each

customer: the maximum or reservation price they are willing to pay

– Second degree price discrimination• Pricing according to quantity consumed –

or in blocks– Third degree price discrimination

• Dividing the market into two or more groups, each having its own demand function and different price elasticities of demand

60ME Slot 2

P*

Q*

Consumer surplus when a single price P* is charged.

Part of producer surplus when a single price P* is charged.

Additional profit from perfect price discrimination, i.e., Deadweight loss being converted into monopoly profit.

Quantity

$/Q Pmax

D = AR

MR

MC

Q**

PC

With perfect discrimination• Each customer pays his reservation price•Profits increase

Additional Profit From Perfect First-Degree Price Discrimination

61ME Slot 2

Caveats to First Degree Price Discrimination

• In practice, transactions costs and information constraints make this difficult to implement perfectly (but car dealers and some professionals come close).

• Price discrimination won’t work if consumers can resell the good.

62ME Slot 2

Implementing First-Degree Price

Discrimination• Question: Why would a producer have

difficulty in achieving first-degree price discrimination?

• Answer: 1) Too many customers (impractical); 2) Could not estimate the reservation price for each customer– Examples of imperfect price discrimination

where the seller has the ability to segregate the market to some extent and charge different prices for the same product:• Lawyers, doctors, accountants

• Car salesperson (15% profit margin)

• Colleges and universities63ME Slot 2

Second-Degree Price Discrimination

Quantity

$/Q

D

MR

MC

AC

P0

Q0

Without discrimination: P = P0 and Q = Q0. With second-degree

discrimination there are threeprices P1, P2, and P3.(e.g. electric utilities)

P1

Q1

1st Block

P2

Q2

P3

Q3

2nd Block 3rd Block

Second-degree pricediscrimination is pricing

according to quantityconsumed--or in blocks.

What happens to deadweight loss?

64ME Slot 2

Third Degree Price Discrimination

1) Divides the market into groups.

2) Each group has its own demand function.

Most common type of price discrimination:• Examples: airlines, liquor, vegetables,

discounts to students and senior citizens.

• Third-degree price discrimination is feasible when the seller can separate his/her market into groups who have different price elasticities of demand (e.g. business air travelers versus vacation air travelers) and there is no seepage across those groups

65ME Slot 2

•MC = MR1 = P1(1+1/E1) = MR2 = P2(1+1/E2)

•=> P1/P2 = (1+1/E2)/(1+1/E1)

•=> Pricing: Charge higher price to group with a lower demand elasticity

Relative prices underThird-degree price

discrimination

66ME Slot 2

Third-Degree Price Discrimination

Quantity

D2 = AR2

MR2

$/Q

D1 = AR1MR1

MRT

MC

P1

P2

Q1

Q2 Qt

How do you get MRT from MR1 and MR2? Through horizontal orvertical addition?

67ME Slot 2

An Example

• Suppose the elasticity of demand for Kodak film in the US is EU = -1.5, and the elasticity of demand in Japan is EJ = -2.5.

• Marginal cost of manufacturing film is $3.

• PU = [EU/(1+ EU)] MC = [-1.5/(1 - 1.5)] $3 = $9

• PJ = [EJ/(1+ EJ)] MC = [-2.5/(1 - 2.5)] $3 = $5

• Kodak’s optimal third-degree pricing strategy is to charge a higher price in the US, where demand is less elastic.

68ME Slot 2

Further Insights: No Sales to Smaller Market under Third Degree Price

Discrimination

Quantity

D2

MR2

$/Q

MC

D1

MR1 Q*

P*

Group one, with demand D1, is not willing to pay enough for the good to

make price discri-mination profitable.

69ME Slot 2

Case-let: The Economics of Coupons and Rebates

• Those consumers who are more price elastic will tend to use the coupon/rebate more often when they purchase the product than those consumers with a less elastic demand.

• Coupons and rebate programs allow firms to price discriminate.

70ME Slot 2

Price Elasticities of Users/ Nonusers of Coupons

Product Non-users Users (larger in magnitude)

Toilet tissue -0.60 -0.66

Stuffing/dressing -0.71 -0.96

Shampoo -0.84 -1.04

Cooking/salad oil -1.22 -1.32

Dry mix dinner -0.88 -1.09

Cake mix -0.21 -0.43

Cat food -0.49 -1.13

Frozen entrée -0.60 -0.95

Gelatin -0.97 -1.25

Spaghetti sauce -1.65 -1.81

Crème rinse/conditioner -0.82 -1.12

Soup -1.05 -1.22

Hot dogs -0.59 -0.77

71ME Slot 2

Economics of Coupons and Rebates (continued)

• Elasticity of demand for Pillsbury cake mix VS. all cake mix

– Users of coupons: -4 (Pillsbury) -0.43 (all)

– Nonusers: -2 (Pillsbury) -0.21 (all)

Thus, PE for Pillsbury 8 to 10 times PE for all cake mix

• Using:

• Price of non-users should be 1.5 times users’

– Or, if cake mix sells for $1.50, coupons should be 50 cents

)11(

)11(

1

2

2

1

E

E

P

P

72ME Slot 2

Elasticities of Demand for Air Travel

Fare CategoryElasticity First-Class Unrestricted Coach Discount

ticket

Price -0.3 -0.4 -0.9

Income 1.2 1.2 1.8

73ME Slot 2

Case-let on Airline Fares

• Differences in elasticities imply that some customers will pay a higher fare than others.

• Business travelers have few choices and their demand is less elastic.

• Casual travelers have choices and are more price sensitive.

• The airlines separate the market by setting various restrictions on the tickets.

– Less expensive: notice, stay over the weekend, no refund– Most expensive: no restrictions

74ME Slot 2

Further Insights: A possible case of ‘dumping’ in global market

Price etc.

ARd

MRd

ARw = MRw

MRt

ACMC

QuantityQd

Pd

Qt

Does opening of a global competitive market help the monopolist,and if so, how? Can it be brought under purview of ‘anti-dumpingmeasures’?

75ME Slot 2

Further Insights: Can discriminating monopoly be socially useful?

Price etc.

AR1

MR1

ACMC

QuantityQ2

P2

While a monopolist in either market can’t work given higher costs,a discriminating monopolist across the two markets can make ∏>0, due to economies of scale, when two markets are combined.

MR2

CMRAR2

CAR

P1

Q1

∏>0

76ME Slot 2

Two-Part Pricing• When it isn’t feasible to charge

different prices for different units sold, but demand information is known, two-part pricing may permit you to extract all surplus from consumers.

• Two-part pricing consists of a fixed fee and a per unit charge.– Example: Athletic club memberships.

77ME Slot 2

How Two-Part Pricing Works

1. Set price at marginal cost.2. Compute consumer

surplus.3. Charge a fixed-fee equal

to consumer surplus.

Quantity

D

10

8

6

4

2

1 2 3 4 5

MC

Fixed Fee = Profits* = $16

Price

Per UnitCharge

* Assuming no fixed costsME Slot 2

Block Pricing• The practice of packaging multiple

units of an identical product together and selling them as one package.

• Examples– Paper.– Six-packs of soda.– Different sized of cans of green beans.

79ME Slot 2

An Algebraic Example

• Typical consumer’s demand is P = 10 - 2Q

• C(Q) = 2Q• Optimal number of units in a package?• Optimal package price?

80ME Slot 2

Optimal Quantity To Package: 4 Units

Price

Quantity

D

10

8

6

4

2

1 2 3 4 5

MC = AC

81ME Slot 2

Optimal Price for the Package: $24

Price

Quantity

D

10

8

6

4

2

1 2 3 4 5

MC = AC

Consumer’s valuation of 4units = .5(8)(4) + (2)(4) = $24Therefore, set P = $24!

82ME Slot 2

Costs and Profits with Block Pricing

Price

Quantity

D

10

8

6

4

2

1 2 3 4 5

MC = AC

Profits* = [.5(8)(4) + (2)(4)] – (2)(4)= $16

Costs = (2)(4) = $8

* Assuming no fixed costs 83ME Slot 2

Inter-temporal Price Discrimination

• Separating the Market With Time:– When a product is initially released, its

demand is inelastic• Movie• Book • Computer

– Once the market has yielded a maximum profit, firms lower the price to appeal to the general market with a more elastic demand• Paper back books• Dollar Movies• Discount computers 84ME Slot 2

Inter-temporal Price Discrimination

Quantity

AC = MC

$/Q

Over time, demand becomesmore elastic and price

is reduced to appeal to the mass market.

Q2

MR2

D2 = AR2

P2

D1 = AR1MR1

P1

Q1

Consumers are dividedinto groups over time.

Initially, demand is lesselastic resulting in a

price of P1 .

Note MC need not be the same over time; Hence MR1 need not be equal to MR2.

85ME Slot 2

Distinctive features of peak load pricing

• Demand for some products may peak at particular times, e.g., rush hour traffic, electricity consumption in summer afternoon

• Markets may be separated on basis of time. During peak demand time, capacity restraints will increase MC. Increased MC would indicate a higher price

• MR is not equal for each market because one market does not impact the other market

86ME Slot 2

MR1

D1 = AR1

MC

P1

Q1

Peak-load price = P1 .

Peak-Load Pricing (when further investment in capacity not needed)

Quantity

$/Q

MR2

D2 = AR2

Off- load price = P2 .

Q2

P2

87ME Slot 2

Capacity Determination under Peak Load Pricing

Model

bB

P1

P2

X*=X2

D2

DT

D1

MRT

MR2

MR1

MC

MR1 MR2

Prices etc.

Capacity

How do you get MRT in this situation – through vertical or horizontal addition of MR1 & MR2?

X1

88ME Slot 2

Commodity Bundling

• The practice of bundling two or more products together and charging one price for the bundle.

• Examples– Vacation packages.– Computers and software.– Film and developing.

89ME Slot 2

Conditions necessary for

bundling

–Heterogeneous customers–Price discrimination is not possible

–Demands must be negatively correlated

90ME Slot 2

Bundling Example With Two Consumers (theaters A-

B)

Spiderman Spaceballs

Theater A $12,000 $3,000

Theater B $10,000 $4,000

• Renting the movies separately would result in each theater paying the lowest reservation price for each movie

– Total Revenue = $26,000

• If the movies are bundled and if each were charged the lower of the two prices

– Total revenue will be $28,000.

Reservation Price

91ME Slot 2

Bundling Example With Two Consumers – Importance of Negative

Correlation of Demands• If the demands were positively correlated

(Theater A would pay more for both films as shown), bundling would not result in an increase in revenue.

Gone with the Wind Getting Gertie’s Garter

Theater A $12,000 $4,000

Theater B $10,000 $3,000

If the movies are bundled and if each were charged the lower of the two prices, total revenue will be $26,000, the same as by selling the films, separately.

92ME Slot 2

An Example that Illustrates Kodak’s

Moment• Total market size for film and developing is 4

million consumers.• Four types of consumers

– 25% will use only Kodak film (F).– 25% will use only Kodak developing (D).– 25% will use only Kodak film and use only

Kodak developing (FD).– 25% have no preference (N).

• Zero costs (for simplicity).• Maximum price each type of consumer will pay

is as follows:

93ME Slot 2

Reservation Prices for Kodak Film and Developing

by Type of Consumer

Type Film DevelopingF $8 $3

FD $8 $4D $4 $6N $3 $2

94ME Slot 2

Optimal Film Price?Type Film Developing

F $8 $3FD $8 $4D $4 $6N $3 $2

Optimal Price is $8; only types F and FD buy resulting in profits of $8 x 2 million = $16 Million.

At a price of $4, only types F, FD, and D will buy (profits of $12 Million).

At a price of $3, all will types will buy (profits of $12 Million).

95ME Slot 2

Optimal Price for Developing?

Type Film DevelopingF $8 $3

FD $8 $4D $4 $6N $3 $2

Optimal Price is $3, to earn profits of $3 x 3 million = $9 Million.

At a price of $6, only “D” type buys (profits of $6 Million).

At a price of $4, only “D” and “FD” types buy (profits of $8 Million).

At a price of $2, all types buy (profits of $8 Million).

96ME Slot 2

Total Profits by Pricing Each Item Separately?

Type Film DevelopingF $8 $3

FD $8 $4D $4 $6N $3 $2

Total Profit = Film Profits + Development Profits = $16 Million + $9 Million = $25 Million

Surprisingly, the firm can earn even greater profits by bundling!

97ME Slot 2

Pricing a “Bundle” of Film and Developing

Type Film Developing Value of BundleF $8 $3 $11

FD $8 $4 $12D $4 $6 $10N $3 $2 $5

98ME Slot 2

What’s the Optimal Price for a Bundle?

Type Film Developing Value of BundleF $8 $3 $11

FD $8 $4 $12D $4 $6 $10N $3 $2 $5

Optimal Bundle Price = $10 (for profits of $30 million)

99ME Slot 2

Cross-Subsidies• Prices charged for one product are

subsidized by the sale of another product.

• May be profitable when there are significant demand complementarities effects.

• Examples– Browser and server software.– Drinks and meals at restaurants.

100ME Slot 2

Tying• Practice of requiring a customer to purchase

one good in order to purchase another.• Allows the seller to meter the customer and

use a two-part tariff to discriminate against the heavy user

• Examples:– Xerox machines and the paper

– IBM mainframe and computer cards

– Renting out tractor along with driver

– Rural moneylenders providing credit against sale of output and/or input purchase (even land leasing-in) contract

• Is tying necessarily ‘exploitative’?

101ME Slot 2

Meaning of Two-Part Tariff

• The purchase of some products and services can be separated into two decisions, and therefore, two prices. Examples

1) Amusement Park• Pay to enter• Pay for rides and food within the park

2) Tennis Club• Pay to join• Pay to play

• Pricing decision is setting the entry fee (T) and the usage fee (P), thus choosing the trade-off between free-entry and high use prices, or

high-entry and zero use prices

102ME Slot 2

Usage price P* is set where MC = D. Entry price T* is equal to the entire consumer surplus. How do you find out T* ?

T*

Two-Part Tariff with a Single Consumer

Quantity

$/Q

MCP*

D

103ME Slot 2

D2 = consumer 2

D1 = consumer 1

Two-Part Tariff with Two Consumers

Quantity

$/Q

MC

Q1Q2

The price, P*, will be greater than MC. Set T* at the surplus value of D2.T*

P*

Thus there is a trade-off between high entry fee & high user price

A

C

Π=2T*+(P*-MC).(Q1+Q2)>ΔABC

B

104ME Slot 2

The Two-Part Tariff with Many Consumers

• No exact way to determine P* and T*.• Must consider the trade-off between the entry

fee T* and the use fee P*.– Low entry fee=> High sales revenue, but less

entry,

– Low price=> More use, but falling profit with lower price.

• To find optimum combination, choose several combinations of P and T

• Choose the combination that maximizes profit

• Rule of Thumb– Similar demand: Choose P close to MC and high

T– Dissimilar demand: Choose high P and low T.

105ME Slot 2

Two-Part Tariff With A Twist

• Suppose, entry price (T) entitles the buyer to a certain number of free units

•Gillette razors with several blades

•Amusement parks with some tokens

•On-line with free time

106ME Slot 2

PRICING PRICING STRATEGIES FOR STRATEGIES FOR

FIRMS WITH FIRMS WITH MARKET POWER-IIMARKET POWER-II

PRICING PRICING STRATEGIES FOR STRATEGIES FOR

FIRMS WITH FIRMS WITH MARKET POWER-IIMARKET POWER-II

ME, SESSION 13ME, SESSION 13

1717thth August, 2010 August, 2010

PROF. SAMAR K. DATTAPROF. SAMAR K. DATTA

Checklist for students• Pricing in Markets with Intense Price

Competition– Price Matching – Brand Loyalty– Randomized Pricing

• Advertising• Pricing of Joint Products• Pricing for Special Cost and Demand

Structures Problem of Double Marginalization & Transfer Pricing– Importance & significance of transfer pricing in vertically

integrated M-form of firms– Transfer pricing with no outside market (Fig 1)– Transfer pricing with a competitive outside market– Elaborating concepts with a numerical example from P & R

108ME Slot 2

Pricing in Markets with Intense Price Competition

• Price Matching– Advertising a price and a promise to match any lower price

offered by a competitor.– No firm has an incentive to lower their prices.– Each firm charges the monopoly price and shares the

market.• Induce brand loyalty

– Some consumers will remain “loyal” to a firm; even in the face of price cuts by a rival firm.

– Advertising campaigns and “frequent-user” style programs can help firms induce loyalty among consumers.

• Randomized Pricing– A strategy of constantly changing prices.– Decreases consumers’ incentive to shop around as they

cannot learn from experience which firm charges the lowest price.

– Reduces the ability of rival firms to undercut a firm’s prices.

109ME Slot 2

Advertising

• Assumptions– Firm sets only one price– Firm knows Q(P,A)

• How quantity demanded depends on price and advertising

110ME Slot 2

Q0

0

P0

Q1

1

P1

AR

MR

AR and MR are averageand marginal revenue whenthe firm doesn’t advertise.

MC

If the firm advertises, its average and marginalrevenue curves shift to

the right -- average costsrise, but marginal cost

does not.

AR’

MR’

AC’

Effects of Advertising

Quantity

$/Q

AC

111ME Slot 2

Advertising• Choosing Price and Advertising

Expenditure

• A Rule of Thumb for Advertising

adv. of MC full1

)(),(

A

QMC

A

QPMR

AQCAPPQ

Ads

ratio sales toAdv.

1)(

pricingfor /1/)(

PQ

A

A

Q

Q

A

P

MCP

A

QP-MC

EPMCP P

112ME Slot 2

Advertising• A Rule of Thumb for Advertising

• To maximize profit, the firm’s advertising-to-sales ratio should be equal to minus the ratio of the advertising and price elasticities of demand

Thumb of Rule

demand of elasticity Adv.

P

)(

1)(

))((

PA

A

EEPQA

EPMCP

EAQQA

113ME Slot 2

Advertising – An Example• R(Q) = $1 million/yr• $10,000 budget for A (advertising--1% of

revenues)• EA = 0.2 (increase budget $20,000, sales

increase by 20%• EP = -4 (markup price over MC is substantial)

• Should the firm increase advertising?

• YES– A/PQ = -(0.2/-4) = 5%– Increase budget to $50,000

114ME Slot 2

Meaning of Joint Products

• Goods jointly produced in fixed proportions:– Interdependence in production– Single marginal cost curve for both

products or product package; e.g., beef & hides

• However, demand curves & MR curves are independent

• Pricing decision must recognize inter-dependence in production– Marginal revenue of product package is

vertical sum of two MR curves;

115ME Slot 2

Pricing of Joint Products w/o Excess production of

Hides

PH

PB

DB

MRT

DH

MRT

MRB

MRH

MRH MRB

Prices etc.

QuantityX*

MC

116ME Slot 2

Pricing of Joint Products with Excess production of

Hides

PH

PB

DB

MRT

DH

MRT

MRB

MRH

MRH MRB

Prices etc.

QuantityXH

MC

XB 117ME Slot 2

Problem of Double Marginalization• Consider a large firm with two divisions:

– the upstream division is the sole provider of a key input.– the downstream division uses the input produced by the

upstream division to produce the final output.

• Incentives to maximize divisional profits leads the upstream manager to produce where MRU = MCU.– Implication: PU > MCU.

• Similarly, when the downstream division has market power and has an incentive to maximize divisional profits, the manager will produce where MRD = MCD.– Implication: PD > MCD.

• Thus, both divisions mark price up over marginal cost resulting in in a phenomenon called double marginalization.– Result: less than optimal overall profits for the firm.

118ME Slot 2

Notion of Transfer Pricing

• To overcome double marginalization, the internal price at which an upstream division sells inputs to a downstream division should be set in order to maximize the overall firm profits.

• To achieve this goal, the upstream division produces such that its marginal cost, MCu, equals the net marginal revenue to the downstream division (NMRd):

NMRd = MRd - MCd = MCu119ME Slot 2

Importance & Significance of Transfer Pricing

• Transfer price = Price for inter-divisional transaction in a multi-divisional company, which is a major determinant of the overall financial performance of the company

• Unless the right transfer price is chosen, the company shall end up having less than maximum profit

• Assuming a competitive external market exists for sale/purchase of the intermediate good, choice of any transfer price other than the competitive outside price shall lead to lower profit

120ME Slot 2

Fig 1: Transfer Pricing When There is No Outside Market

Quantity

(MR – MCA)

MCE

AR

MCA

MR

QA = QE

PE

PA

NMRE

MCA is the marginal cost of assembling cars given the engines. Since one car requires one engine, the marginal product of engines is 1. Therefore the curve (MR – MCA ) is the net marginal revenue curve NMRE for engines. The transfer price PE correctly values the engines used to produce the cars. The same result is obtained if we take intersection point between MR and ∑MC.

∑MC

121ME Slot 2

A Numerical Example

• Demand for automobiles: P = 20,000 – Q• MR for automobiles: MR = 20,000 – 2Q• Down-stream’s assembling cost:

CA(Q) = 8000Q => MCA = 8000

• Up-stream’s cost of production of engines: CE(QE) = 2QE

2 => MCE(QE) = 4QE

• Because of 1-1 correspondence: QE = Q

122ME Slot 2

Numerical Example: Case 1 – No outside market

•NMRE = MR – MCA = 20,000 – 2Q – 8,000 = 12,000 – QE

•NMRE = MCE => 12,000 - 2QE = 4QE

=> QE = 2000

•PE = MCE = 4QE = $8,000

123ME Slot 2

Numerical Example: Case 2 – Outside competitive market

for engines• Let PE,M = 6000 < Transfer price = 8000 =>

Buying some engines from outside• => NMRE = 6000 => 12,000 - 2QE = 6000 2QE

= 6000 => QE = 3000• Company produces more cars & buys some

engines at lower price, given lower cost of engines in outside market

• Up-stream production: MCE = 6000 => 4QE

=6000 => QE = 1500 => Rest 1500 are bought from market

124ME Slot 2

Pricing by Firms Pricing by Firms with Market Power-with Market Power-

IIIIII

Pricing by Firms Pricing by Firms with Market Power-with Market Power-

IIIIII

ME, Session 14ME, Session 14

!8!8thth August, 2010 August, 2010

PROF. SAMAR K. DATTAPROF. SAMAR K. DATTA

Checklist for students

I. Limit Pricing to Prevent EntryII. Predatory Pricing to Lessen CompetitionIII. Raising Rivals’ Costs to Lessen

CompetitionIV. Price Discrimination as a Strategic ToolV. Changing the Timing of DecisionsVI. Penetration Pricing to Overcome Network

Effects

126ME Slot 2

Limit Pricing

• Strategy where an incumbent (existing firm) prices below the monopoly price in order to keep potential entrants out of the market.

• Goal is to lessen competition by eliminating potential competitors’ incentives to enter the market.

127ME Slot 2

Limit Pricing• Incumbent produces

QL instead of monopoly output (QM).

• Resulting price, PL, is lower than monopoly price (PM).

• Residual demand curve is the market demand (DM) minus QL

.• Entry is not profitable

because entrant’s residual demand lies below AC for output<Q.

• Optimal limit pricing results in a residual demand such that, if the entrant entered and produced Q units, its profits would be zero.

Quantity

$

Q M

P M

AC

Entrant's residualdemand curve

D MP = AC

P L

Q LQ

128ME Slot 2

QM

General Diagram for Entry Limiting Pricing

Quantity

$/QMC

AC

D=AR

MR

Pm

QL

PL

QL’

AB

CD

E

With monopoly price-quantity combination (Pm, Qm), Π=PmADC. If limit price is fixed at PL, Π=PLBCE and Q=QL. Now, super-normal profit is still positive but less. Now a new entrant with Q<QL’ will face loss, thus P=PL acts as a barrier to small-scale entry.

129ME Slot 2

Potential Problems with Limit Pricing

• It isn’t generally profitable for the incumbent to maintain an output of QL once entry occurs.

• Rational entrants will realize this and enter.• Solution: Incumbent must link its pre-entry

price to the post-entry profits of the potential entrant.

• Possible links: Commitments by incumbents. Learning curve effects. Incomplete information. Reputation effects.

130ME Slot 2

Potential Problems with Limit Pricing (Continued)

• Even if a link can be forged, it may not be profitable to limit price! Limit pricing is profitable only if the present value of the benefits of limit pricing exceed the up front costs:

.

L D

M L

i

131ME Slot 2

Implications of limit pricing

Π

Time (t)

Discounted profit stream under limit pricing

Discounted profit stream under usual profit-maximization

T

The blue or red path will be chosen depending upon whether t>T or t<T

Higher discount rate would favor the red path, while a lower one would favor the blue one.

132ME Slot 2

Predatory Pricing

• Strategy of pricing below marginal cost to drive competitors out of business, then raising price to enjoy the higher profits resulting from lessened competition.

• Goal is to lessen competition by eliminating existing competitors.

133ME Slot 2

Potential Problems with Predatory Pricing

• Counter strategies: Stop production. Purchase from the predator at the reduced

price and stockpile until predatory pricing is over.

• Rivals can sue under the Sherman Act But it is often difficult for rivals to prove their

case.• Upfront losses incurred to drive out rivals may

exceed the present value of future monopoly profits.

• Predator must have deeper pockets than prey.

134ME Slot 2

Raising Rivals’ Costs

• Strategy where a firm increases the marginal or fixed costs of rivals to distort their incentives.

• Not always profitable, but may be profitable as the following example shows.

135ME Slot 2

Raising a Rival’s Marginal Cost

• Cournot duopoly.• Initial equilibrium at

point A.• Firm 1 raises the

marginal cost of Firm 2, moving equilibrium to point B.

• Firm 1 gains market share and profits.

A

B

Q 2

Q 1

136ME Slot 2

Other Strategies to Raise Rivals’ Costs

• Raise fixed costs in the industry.• If vertically integrated, increase input

prices in the upstream market. Vertical Foreclosure: Integrated firm

charges rivals prohibitive price for an essential input.

The Price-Cost Squeeze: Integrated firm raises input price and holds the final product price constant.

137ME Slot 2

Price Discrimination as a Strategic Tool

• Price discrimination permits a firm to “target” price cuts to those consumers or markets that will inflict the most damage to the rival (in the case of predatory pricing) or potential entrants (in the case of limit pricing).

• Meanwhile, it can continue to charge the monopoly price to its other customers.

• Thus, price discrimination may enhance the value of other pricing strategies.

138ME Slot 2

Changing the Timing of Decisions or the Order of

Moves

• Sometimes profits can be enhanced by changing the timing of decisions or the order of moves. When there is a first-mover

advantage, it pays to commit to a decision first.

When there is a second-mover advantage, it pays to let the other player move first.

139ME Slot 2

Examples of Games with First and Second-Mover

Advantages

• Example 1: Player naming the smaller natural number gets $10, the other players get nothing.– First-mover always earns $10.

• Example 2: Player naming the larger natural number gets $10, the other players get nothing.– Last-mover always earns $10

• Practical Examples?140ME Slot 2

If Firms A and B Make Production Decisions Simultaneously

• Firm A earns $10 by playing its dominant strategy, which is “Low Output.”

Firm A

Firm B

Strategy Low Output High Output

Low Output

High Output

$30, $10 $10, $15

$20, $5 $1, $2

141ME Slot 2

But if A Moves First:• Firm A can earn

$20 by producing a high output!

• Requires Commitment to a

high output. Player B

observes A’s commitment prior to making its own production decision.

Low Output ($30, $10)

Low Output

High Output ($10, $15)

Low Output ($20, $5)

High Output

High Output ($1, $2)

A

B

B

142ME Slot 2

Networks• A network consists of links that connect

different points in geographic or economic space.

• One-way Network – Services flow in only one direction. Examples: water, electricity.

• Two-way Network – Value to each user depends directly on how many other people use the network. Examples: telephone, e-mail.

143ME Slot 2

Example: A Two-Way Star Network Linking 7 Users

• Point H is the hub.

• Points C1 through C7 are nodes representing users.

• Total number of connection services is n(n - 1) = 7(7-1) = 42.

H C1

C7

C6

C5

C3

C4 C2

144ME Slot 2

Network Externalities• Direct Network Externality – The direct

value enjoyed by the user of a network because other people also use the same network.

• Indirect Network Externality – The indirect value enjoyed by the user of a network because of complementarities between the size of the network and the availability of complementary products or services.

• Negative Externalities such as congestion and bottlenecks can also arise as a network grows.

145ME Slot 2

Penetration Pricing to Overcome Network

Effects• Problem: Network externalities

typically make it difficult for a new network to replace or compete with an existing network.

• Solution: Penetration Pricing– The new network can charge an initial price

that is very low, give the product away, or even pay consumers to try the new product to gain users.

– Once a critical mass of users switch to the new network, prices can be increased.

146ME Slot 2

The New Network GameWithout Penetration Pricing

• Coordination Problem Neither user has an incentive to unilaterally switch to H2, even though both users

would benefit if they simultaneously switched. With many users, it is difficult to coordinate a move to the better equilibrium. Users may stay locked in at the red equilibrium instead of moving to green one.

Network Provider H1 H2

H1

H2

User 2

User 1 $0, $0$10, $10

$20, $20$0, $0

Table 13-2 A Network Game

147ME Slot 2

The New Network GameWith Penetration Pricing

Network Provider H1 H1 & H2

H1

H1 & H2

User 2

User 1 $10, $11$10, $10

$21, $21$11, $10

Table 13-3 The Network Game with Penetration Pricing

• Network provider H2 pays consumers $1 to try its network; consumers have nothing to lose in trying both networks. The green cell is the equilibrium.

• Users will eventually realize that H2 is better than H1 and that other users have access to this new network.

• Users will eventually quit using H1, at which point provider H2 can eliminate $1 payment and start charging for network access.

148ME Slot 2

Conclusion• A number of strategies may enhance

profits: Limit pricing. Predatory pricing. Raising rivals’ costs. Exercising first- or second-mover

advantages. Penetration pricing.

• These strategies are not always the best ones, though, and care must be taken when using any of the above strategies.

149ME Slot 2

ME Session 15ME Session 15August 23, 2010August 23, 2010Markets for Factor Markets for Factor

Inputs-IInputs-IProf. Samar K. DattaProf. Samar K. Datta

Markets for Factor Markets for Factor Inputs-IInputs-I

Prof. Samar K. DattaProf. Samar K. Datta

Overview• Typology of Situations to be covered

• Equilibrium in Competitive Factor Markets– Characteristics of Competitive Factor Markets

– Demand for Factor Input with Only One Variable Factor

– Distinguishing between Marginal Revenue Product (MRPL) & Value of Marginal Product (VMPL) of Labor

– Comparing Input and Output Market Equilibrium Conditions

– Firm’s Demand Curve for Labor (with Variable Capital)

– Industry Demand for Labor

– A Firm’s Input Demand in a Competitive Factor Market

– Labor Market Equilibrium under Competition & Monopoly

– Economic Rent

151ME Slot 2

Typology of Situations

Typology 1 2 3 4 5 6 7 8

Seller(Product) comp m’ply comp m’ply comp m’ply

comp m’ply

Buyer(Labor) comp comp m’sny m’sny comp comp

m’sny m’sny

Seller(Labor) comp comp comp comp m’ply m’ply

m’ply m’ply

152ME Slot 2

Characteristics of Competitive Factor

Markets

1) Large number of buyers & sellers of the factor of production

2) The buyers and sellers of the factor of production are price takers

153ME Slot 2

Demand for Factor Input with Only One Variable Factor

• Demand for factor inputs is a derived demand, derived from factor cost and output demand.

• Measuring the Value of a Worker’s Output– Marginal Revenue Product of Labor (MRPL)

– MRPL = (MPL)(MR)

• In a competitive product market – MR = P => VMPL=P.MPL

154ME Slot 2

Marginal Revenue Product (MRPL) &

Value of Marginal Product (VMPL) of

Labor

Hours of Work

Wages($ perhour)

VMPL = MPLx P

Competitive Output Market (P = MR)

MRPL = MPL x MR

Monopolistic Output Market

(MR <P)

155ME Slot 2

w* SL

In a competitive labor market, a firm faces a perfectly elastic supply of labor

and can hire as many workers as it wants at w*.

Hiring by a Firm in theLabor Market (with Capital Fixed)

Quantity of Labor

Price ofLabor

VMPL = DL

L*

The profit maximizing firm willhire L* units of labor at the point

where the marginal revenue productof labor is equal to the wage rate.

156ME Slot 2

A Shift in the Supply of Labor (e.g., due to baby boom/ female

entry)

Quantity of Labor

Price ofLabor

w1S1

VMPL = DL

L1

w2

L2

S2

157ME Slot 2

Comparing Input and Output Market Equilibrium

Conditions

• Input market equilibrium condition:– Under monopoly: MRPL = MR.MPL = W => MR =

W/MPL = MC

– Under competition: VMPL = P. MPL = W => P = AR = MR = W/MPL = MC

• So, input market equilibrium condition is the same as the profit-maximization condition in the output market

158ME Slot 2

VMPL1 VMPL2

When two or more inputs arevariable, a firm’s demand for one input

depends on the marginal revenue product of both inputs.

Firm’s Demand Curve for Labor

(with Variable Capital)

Hours of Work

Wages($ perhour)

0

5

10

15

20

40 80 120 160

When the wage rate is $20, A represents one point on the firm’s demand for labor curve. When the wage rate falls to $15, the VMPL curve shifts, generating a new point C on the firm’s demand for labor curve. Thus A and C are on the demand for labor curve, but B is not. Thus, demand curve for labor is more elastic in the presence of another variable factor, which opens up scope for substitution.

DL

A

B

C

159ME Slot 2

• Assuming all firms respond to a lower wage– All firms would hire more workers.– Market supply would increase.– The market price will fall. – The quantity demanded of labor by

the firm will be smaller.– Thus, industry demand curve would

be less elastic as compared to the curve w/o any price fall.

Industry Demand for Labor

160ME Slot 2

VMPL1

Derivation of the Industry Demand for

Labor

Labor(worker-hours)

Labor(worker-hours)

Wage($ perhour)

Wage($ perhour)

0

5

10

15

0

5

10

15

50 100 150 L0 L2

DL1

Horizontal sum ifproduct price

unchanged

120

VMPL2

L1

Industrydemand

curve withprice falling

DL2

Firm Industry

With lower wages, more labor demanded because of substitution across inputs, output rises with demandunchanged, leading to price fall. At lower price, demand curve for labor shifts down, as shown in diagram.Hence, demand for labor rises less at lower wage rate than what would have happened with fixed price.

161ME Slot 2

SMarket Supplyof fabric

A Firm’s Input Demand in aCompetitive Factor Market

Yards ofFabric (thousands)

Yards ofFabric (thousands)

Price($ peryard)

Price($ peryard)

D

Market Demandfor fabric

100

ME = AE10 10

Supply ofFabric Facing Firm

50

Demand for Fabric

VMP

Observations1) The firm is a price taker at $10.2) S = AE = ME = $103) ME = VMP @ 50 units

162ME Slot 2

SL = AE

SL = AE

DL = VMPL DL = MRPL

VMPL =P *MPL

Labor Market Equilibrium (typology 1 & 2)

Number of Workers Number of Workers

Wage WageCompetitive Output Market Monopolistic Output Market

wC

LC

wM

LM

vM

AB

WC

LC

With monopoly, w↓, L ↓, Q ↓

(VM – WM) =‘exploitation of labor under monopoly

163ME Slot 2

Labor Market Equilibrium

• Equilibrium in a Competitive Output Market– DL(VMPL) = SL

– wC = VMPL

– VMPL = (P)(MPL)– Markets are efficient

• Equilibrium in a Monopolistic Output Market– MR < P– MRP = (MR)(MPL)

– Hire LM at wage wM

– vM = marginal benefit to consumers

– wM = marginal cost to the firm

164ME Slot 2

Labor Market Equilibrium

• Equilibrium in a Competitive Output Market– DL(VMPL) = SL

– wC = VMPL

– VMPL = (P)(MPL)– Markets are efficient

• Equilibrium in a Monopolistic Output Market– MR < P– MRP = (MR)(MPL)

– Hire LM at wage wM

– vM = marginal benefit to consumers

– wM = marginal cost to the firm

165ME Slot 2

Total expenditure (wage) paidis rectangle 0w* AL*Economic Rent

Economic rent is ABW*

B

Economic Rent

Number of Workers

Wage

SL = AE

DL = VMPL

w*

L*

A

0

The economic rent associated with theemployment of labor is the excess of wages

paid above the minimum amount neededto hire workers.

166ME Slot 2

EconomicRent

s1

EconomicRent

s2

Land Rent

Number of Acres

Price($ peracre)

Supply of Land

D2

D1

167ME Slot 2

ME Session 16: ME Session 16: August 24, 2010August 24, 2010Markets for Factor Markets for Factor

Inputs-IIInputs-IIProf. Samar K. DattaProf. Samar K. Datta

Markets for Factor Markets for Factor Inputs-IIInputs-II

Prof. Samar K. DattaProf. Samar K. Datta

Overview• Factor Markets with Monopsony Power

• Factor Markets with Monopoly Power

• Wage discrimination across unionized & non-unionized labor (optional)

• Bilateral Monopoly (monopolist seller of labor facing a monopsonist buyer)

• The Decline of Private Sector Unionism in USA

169ME Slot 2

Factor Markets with Monopsony Power

• Assume– The output market is perfectly competitive.– Input market is pure monopsony.

• Monopsonist forcing wage determination on seller’s supply curve,i.e., forcing labor to get only its minimum supply price.

• Examples of Monopsony Power– Government

• Soldiers• Missiles• B2 Bombers

– NASA• Astronauts

– Company town

170ME Slot 2

SL = Average Expenditure (AE)

MarginalExpenditure (ME)

Why is marginal expendituregreater than SL?

D = VMPL

Marginal and Average Expenditure under monopsony (typology 3:

Competitive seller)

Units of Input

Price(per unitof input)

0 1 2 3 4 65

5

10

15

20

wM = 13

LM

wc

Lc

C

WM

WM< WC

LM< LC

Extent of monopsonisticexploitation

171ME Slot 2

SL = Average Expenditure (AE)

MarginalExpenditure (ME)

Why is marginal expendituregreater than SL?

D = VMPL

Marginal and Average Expenditure under monopsony (typology 4:

monopolist seller)

Units of Input

Price(per unitof input)

0 1 2 3 4 65

5

10

15

20

WM = 13

L*

wc

Lc

C

D=MPL.MRLMM LM

WMM

WMM< WM

LMM< LM

M=monopsonistMM=monopolistcum monopsonist

Extent of exploitation

172ME Slot 2

Factor Markets with Monopoly Power & Alternative Objectives of

Unions

• Just as buyers of inputs can have monopsony power, sellers of inputs can have monopoly power.

• The most important example of monopoly power in factor markets involves labor unions.

• Unions have one of the following three objectives– Maximizing wage rate (WM, LM), M=monopoly situation; – Maximizing wage bill (W2, L2);– Maximizing employment (WC, LC); C=competitive situation

173ME Slot 2

EconomicRent

w1

L1

The quantity of labor L1 that maximizesthe rent that employees earn is determinedby the intersection of the marginal revenueand supply or labor curves; union members

receive a wage rate of w1.

SL

DL

MR

Monopoly Power of Sellers of Labor (typology 5)

Number of Workers

Wageper

worker

A

L2

w2

Finally, if the union wishes to maximize totalwages paid to workers, it should allow L2

union members to be employed at a wagerate of w2 because the marginal revenue

to the union will then be zero.

L*

w*

WM=

WC=WM> WC

LM< LC

LM= LC=

B

C

Note exploitation of labor=0 at A, B, C

174ME Slot 2

EconomicRent

w1

L1

M=only monopolist seller of labor, indicated by points A, B, C

SL

DL

MR

Monopoly Power of Sellers of Labor coupled with Producer Monopoly

Power (typology 6)

Number of Workers

Wageper

worker

A

L2

w2

MM=monopolist seller of labor facing monopolist producer, Indicated by points A’, B’, C’

L*

w*

D’L for monopolist

MR’

WC=

WM=

LMM =LM

WMM< WM

LMM< LM

WMMB

C

A’

B’

C’

175ME Slot 2

Implications of monopoly sale of labor • Seller of labor forcing wage

determination along the demand curve of labor (i.e., trying to realize the full demand price of labor), depending upon whether the seller of product is a competitor (demand for labor being VMPL) or a monopolist (demand for labor being MRPL)

176ME Slot 2

• A Two-Sector Model of Labor Employment– Union monopoly power impacts the

non-unionized part of the economy.

Factor Markets with Monopoly Power

177ME Slot 2

Wage Determination/discrimination inUnionized and Non-unionized Sectors

(optional)

MCU

MCNU

D

DUMRU

WC

WU

LCLU

D is total demand for labor, assuming that both union and non-union labor are physically identical

DU is demand for union labor, which due to the union activities takes precedence over demand for non-union labor

LC is total labor employed (sum of union i.e. LU and non-union labor i.e. LC - LU)

D’D is the portion of total labor demand that has to be satisfied by non-union labor

D’

Additional wage bill due to union

To see how union monopoly power impacts the non-unionized part of the economy

178ME Slot 2

Bilateral Monopoly: Market in which a Monopolist seller (MP) of labor sells to a

Monopsonist buyer (MS) of labor (typology 7)

Numberof Workers

Wageper

worker

DL = VMPL

MR

5

10

15

20

25

10 20 40

SL = AE

ME

25

19

WagePossibilities

wC

WMP=

WMS=

179ME Slot 2

Implications of Bilateral Monopoly for Wages

• Monopolist seller of labor will try to set wage rate at WMP.

• Monopsonist buyer of labor will try to set wage rate at WMS.

• Depending upon the relative bargaining power of the buyer and seller of labor, the wage rate will lie between these two extremes.

180ME Slot 2

Bilateral Monopoly: Market in which a Monopolist Factor Supplier sells to a

Monopsonist cum Monopolist (typology 8: Optional)

Numberof Workers

Wageper

worker

DL = VMPL

DL’=MRPL

5

10

15

20

25

10 20 40

SL = AE

ME

25

19

WagePossibilities

wC

WMP=

WMS=

MR’L

W’MP

W’MS

W’MP< WMP

L’MP< LMP

W’MS< WMS

L’MS< LMS

L’MP L’MS

LMP

LMS

181ME Slot 2

Who will win under Bilateral Monopoly?

• The union will win if its threat to strike is credible.

• The firm will win if its threat to hire non-union workers is credible.

• If both make credible threats, the wage will be at Wc.

182ME Slot 2

The Decline of Private Sector Unionism in USA

• Observations– Union membership and monopoly power has

been declining.– Initially, during the 1970’s, union wages relative

to non-union wages fell.– In the 1980’s union wages stabilized relative to

non-union wages.– In the 1990’s membership has been falling and

wage differential has remained stable.• Explanations

– The unions have been attempting to maximize the individual wage rate instead of total wages paid.

– The demand for unionized employees has probably become increasingly elastic as firms find it easier to substitute capital for skilled labor.

183ME Slot 2

Risk & Uncertainty in Risk & Uncertainty in Decision MakingDecision Making

Session 17: August 25, Session 17: August 25, 20062006

Prof. Samar K. DattaProf. Samar K. Datta

Session 17: August 25, Session 17: August 25, 20062006

Prof. Samar K. DattaProf. Samar K. Datta

Topics to be discussed

• Describing Risk

• Preferences Toward Risk

• Reducing Risk

• Demand for Risky Assets

185ME Slot 2

Describing & Interpreting Risk

• To measure risk one must know:

1) All of the possible outcomes.

2) The likelihood that each outcome will occur (its probability).

• Objective Interpretation– Based on the observed frequency of past events

• Subjective Interpretation– Based on perception or experience with or without

an observed frequency– Different information or different abilities to

process the same information can influence the subjective probability

186ME Slot 2

Describing Risk: Expected Value

• The weighted average of the payoffs or values resulting from all possible outcomes.

– The probabilities of each outcome are used as weights

– Expected value measures the central tendency; the payoff or value expected on average

– E(X) = p1x1 + p2x2 + …..+ pnxn

187ME Slot 2

• While the expected values are the same, the variability is not.

• Greater variability from expected values signals greater risk.

• Deviations or differences between expected payoff and actual payoff are important

– The standard deviation measures the square root of the average of the squares of the deviations of the payoffs associated with each outcome from their expected value.

Standard deviation measuring risk

188ME Slot 2

Example of a risky situation

• Job 1 is a job in which the income ranges from $1000 to $2000 in increments of $100 that are all equally likely.

• Job 2 is a job in which the income ranges from $1300 to $1700 in increments of $100 that, also, are all equally likely.

– Job 1: greater spread & standard deviation

– Job 2: peaked distribution - extreme payoffs are less likely

• Decision Making

– A risk avoider would choose Job 2: same expected income as Job 1, but with less risk.

189ME Slot 2

Preferences Toward Risk

• Assumptions for choosing among risky alternatives– Consumption of a single commodity– The consumer knows all probabilities– Payoffs measured in terms of utility– Utility function given

• Defining risk averseness – i.e., preferences toward risk– A person who prefers a certain given income to a

risky income with the same expected value.– A person is considered risk averse if he has a

diminishing marginal utility of income– The use of insurance demonstrates risk aversive

behavior– If an individual gets more utility from his present

lower-paid job than a higher-paid risky job, he is said to be risk averse

190ME Slot 2

Income ($1,000)

Utility The consumer is riskaverse because she

would prefer a certainincome of $20,000 to a

gamble with a 0.5 probabilityof $10,000 and a 0.5

probability of $30,000.

E

10

10 15 20

1314

16

18

0 16 30

AB

C

D

Risk Averse Preferences Toward Risk

191ME Slot 2

Income ($1,000)10 20

Utility

0 30

6A

E

C

12

18

The consumer is riskneutral as he is indifferent

between certain eventsand uncertain events

with the same expected income.

Risk Neutral Preferences Toward Risk

A person is said to be risk neutral if he shows no preference between a certain income, and an uncertain one with the same expected value.

192ME Slot 2

Income ($1,000)

Utility

0

3

10 20 30

A

E

C8

18The consumer is riskloving because she

would prefer the gamble to a certain income.

Risk Loving Preferences

A person is said to be risk loving if he shows a preference toward an uncertain income over a certain income with the same expected value. Examples: gambling, criminal activity

193ME Slot 2

Notion of Risk Premium

• The risk premium is the amount of money that a risk-averse person would pay to avoid taking a risk.

• R = U[E(Y)] – E[U(Y)] in utility terms

• It can also be measured along the horizontal axis in terms of money

• The greater the variability – i.e., the spread of values of the random variable around mean, the greater the risk premium.

• The greater the concavity of the utility curve, the greater the risk premium – it means a much higher weight is assigned to a decline in income as compared to an equal amount of rise in income

194ME Slot 2

Income ($1,000)

Utility

0 10 16

Here , the risk premiumis $4,000 because a

certain income of $16,000gives the person the same

expected utility as the uncertain income that

has an expected value of $20,000.

10

18

30 40

20

14

A

CE

G

20

F

Risk Premium in money terms

Estimating Risk Premium

Risk premium in utility terms

195ME Slot 2

Meaning of High Degree of Risk Aversion

Standard Deviation of Income

ExpectedIncome

Here increase in Standard deviation requires a large increase in income to maintainsatisfaction.

U1

U2

U3

196ME Slot 2

Meaning of Low Degree of Risk Aversion

Standard Deviation of Income

ExpectedIncome

Slightly Risk Averse =>A large increase in standarddeviation requires only a small increase in incometo maintain satisfaction.

U1

U2

U3

197ME Slot 2

Indifference curves of risk-neutral & risk-loving

people• Risk-neutrality means horizontal

indifference curve showing zero premium the consumer is willing to pay to buy hedge against risk (i.e., he doesn’t mind bearing more risk)

• Risk-lover willing to pay (i.e., make sacrifice in terms expected income) to enjoy the thrill of greater risk-bearing – thus making indifference curves usual downward sloping

198ME Slot 2

Risk Reduction Strategies

• Three ways consumers attempt to reduce risk are:

1) Diversification

2) Insurance

3) Obtaining more information

• Firms can reduce risk by diversifying among a variety of activities that are not closely related.

199ME Slot 2

Reducing Risk thro’ Diversification

• Diversification– Reducing risk by allocating resources to a

variety of activities whose outcomes are not closely related

• Example: – Suppose a firm has a choice of selling air

conditioners, heaters, or both– The probability of it being hot or cold is 0.5– How does a firm decide what to sell?

200ME Slot 2

Income from Sales of Appliances

Hot Weather

Cold Weather

Air conditioner sales

$30,000 $12,000

Heater sales

12,000 30,000

201ME Slot 2

Diversification – Example

• If the firm sells only heaters or air conditioners their income will be either $12,000 or $30,000

• Their expected income would be:– 1/2($12,000) + 1/2($30,000) = $21,000

• If the firm divides their time evenly between appliances, their air conditioning and heating sales would be half their original values

• If it were hot, their expected income would be $15,000 from air conditioners and $6,000 from heaters, or $21,000

• If it were cold, their expected income would be $6,000 from air conditioners and $15,000 from heaters, or $21,000

• With diversification, expected income is $21,000 with no risk• Better off diversifying to minimize risk• Firms can reduce risk by diversifying among a variety of

activities that are not closely related

202ME Slot 2

Reducing Risk – The Stock Market

• If invest all money in one stock, then take on a lot of risk– If that stock loses value, you lose all

your investment value

• Can spread risk out by investing in many different stocks or investments– Ex: Mutual funds

203ME Slot 2

Reducing Risk – Insurance

• Risk averse are willing to pay to avoid risk• If the cost of insurance equals the expected loss, risk averse

people will buy enough insurance to recover fully from a potential financial loss

• For the risk averse consumer, guarantee of same income regardless of outcome has higher utility than facing the probability of risk

• Expected utility with insurance is higher than without

• Purchases of insurance transfers wealth and increases expected utility

– Actual Fairness: When the insurance premium = expected payout

204ME Slot 2

Value of Title InsuranceWhen Buying a House : An

Example

• In the absence of title insurance, a risk averse buyer would pay much less for the house

• By reducing risk, title insurance thus increases the value of the house by an amount far greater than the premium.

205ME Slot 2

The Decision to Insure – An example

206ME Slot 2

How does insurance work? The Law of Large

Numbers• Insurance companies know that although single

events are random and largely unpredictable, the average outcome of many similar events can be predicted

• When insurance companies sell many policies, they face relatively little risk

• Insurance companies can be sure total premiums paid will equal total money paid out

• Companies set the premiums so money received will be enough to pay expected losses

207ME Slot 2

Why no insurance against certain events?• Some events with very little probability

of occurrence such as floods and earthquakes are no longer insured privately– Cannot calculate true expected values and

expected losses– Governments have had to create insurance

for these types of events• Ex: National Flood Insurance Program

208ME Slot 2

The Value of Information

• Risk often exists because we don’t know all the information surrounding a decision

• Because of this, information is valuable and people are willing to pay for it

• The value of complete information– The difference between the expected value

of a choice with complete information and the expected value when information is incomplete

209ME Slot 2

The Value of Information – Example

• Per capita milk consumption has fallen over the years• The milk producers engaged in market research to develop

new sales strategies to encourage the consumption of milk • Milk advertising increases sales most in the spring• Allocating advertising based on this information in New York

increased profits by 9% or $14 million• The cost of the information was relatively low, while the

value was substantial (increased profits)• Findings

– Milk demand is seasonal with the greatest demand in the spring

– Price elasticity of demand is negative and small– Income elasticity is positive and large

210ME Slot 2

Reducing Risk: The Value of Information

• Value of Complete Information: The difference between the expected value of a choice with complete information and the expected value when information is incomplete.

• An example:

– Suppose a store manager must determine how many fall suits to order:

– 100 suits cost $180/suit– 50 suits cost $200/suit– The price of the suits is $300– Unsold suits can be returned for half cost.– The probability of selling each quantity is .50.

• Pay-off matrix:Sale of 50 Sale of 100 Expected Profit

Buy 50 suits $5,000 $5,000 $5,000

Buy 100 suits $1,500 $12,000 $6,750

211ME Slot 2

Implications of the Example

• With incomplete information:

– Risk Neutral: Buy 100 suits

– Risk Averse: Buy 50 suits

• Value of information can arise from: (1) reading the market demand correctly, once uncertainty is resolved & placing order accordingly; (2) resolving uncertainty in advance, i.e., ex ante. This caselet is dealing with only (1)

• The expected value with complete information is $8,500.8,500 = .5(5,000) + .5(12,000)

• The expected value with uncertainty (buy 100 suits) is $6,750

• The value of complete information is $1,750, or the difference between the two (the amount the store owner would be willing to pay for a marketing study).

• Is more information always better? Or, can value of information be negative under certain circumstances? (Read carefully Example 5.5 in P&R, p.170)

212ME Slot 2

Economics of Economics of InformationInformation

Economics of Economics of InformationInformation

ME Session 18: August ME Session 18: August 30, 201030, 2010

Prof. Samar K. DattaProf. Samar K. Datta

Overview• Quality Uncertainty and the Market

for Lemons: Problem of Adverse Selection

• Screening & Market Signaling

• Moral Hazard

• The Principal-Agent Problem & Managerial Incentives

• Asymmetric Information in Labor Markets: Efficiency Wage Theory

214ME Slot 2

Quality Uncertaintyand the Market for Lemons

• The lack of complete information when purchasing a used car increases the risk of the purchase and lowers the value of the car.

• The Market for Used Cars:– If buyers and sellers can distinguish

between high and low quality cars, there will be practically two different markets for two related (presumably still imperfectly substitute) products

215ME Slot 2

The Lemons Problem

PH PL

QH QL

SH

SL

DH

DL

5,000

50,000 50,000

The market for high and lowquality cars when buyers and sellers

can identify each car

10,000

DL

DM

DM

75,00025,000

With asymmetric information buyers will find it difficult to determine quality. They lower

their expectations of the average quality ofused cars. Demand for low and high quality

used cars shifts to DM.

DLM

DLM

The increase in QLreduces expectations anddemand to DLM. The adjustment process continues

until demand = DL.for both types of cars

11

2

2

216ME Slot 2

• The Market for Used Cars– With asymmetric information:

• Low quality goods drive high quality goods out of the market.

• The market has failed to produce mutually beneficial trade.

• Too many low and too few high quality cars are on the market.

• Adverse selection occurs; the only cars on the market will be low quality cars.

• More Examples: Gresham’s Law – Bad money (with intrinsic value < face value) throwing out good money

Quality Uncertaintyand the Market for Lemons

217ME Slot 2

Example of Asymmetric Information: The Market for

Medical Insurance • Is it possible for insurance companies to separate

high and low risk policy holders?

– If not, only high risk people will purchase insurance.

– Adverse selection would make medical insurance unprofitable.

• What impact does asymmetric information and adverse selection have on insurance rates and the delivery of automobile accident insurance?

218ME Slot 2

Another Example: The Market for Credit & Role of

Screening

• Asymmetric information creates the potential that only high risk borrowers will seek loans.– Questions:

• How can credit histories help make this market more efficient and reduce the cost of credit?

• Can credit rationing, forcing market to close at below market-clearing price, help?

219ME Slot 2

Implications of Asymmetric Information:

Importance of Reputation and Standardization

• Asymmetric Information and Daily Market Decisions

• Ratings by ‘Better Business Bureau’

• San Francisco’s Forty Niners’ Football Team

• Retail sales

• Antiques, art, rare coins

• Home repairs

• Restaurants

220ME Slot 2

Implications of Asymmetric Information: More Insights

• Question: How can these producers provide high-quality goods when asymmetric information will drive out high-quality goods through adverse selection?– Answer: Reputation

• Question: Why do you look forward to a Big Mac when traveling, even though you would never consider buying one at home?

• Holiday Inn once advertised “No Surprises” to address the issue of adverse selection.

221ME Slot 2

Market Signaling• The process of sellers using signals to

convey information to buyers about the product’s quality helps buyers and sellers deal with asymmetric information.

• Strong Signal– To be effective, a signal must be easier

for high quality sellers to give than low quality sellers.

– Example• Highly productive workers signal with

educational attainment level.

222ME Slot 2

A Simple Model of Job Market Signaling

• Assume Two groups of workers– Group I: Low productivity--AP & MP = 1– Group II: High productivity--AP & MP = 2– The workers are equally divided between Group

I and Group II--AP for all workers = 1.5

• Assume Competitive Product Market– P = $10,000– Employees average 10 years of employment– Group I Revenue = $100,000 (10,000/yr. x 10)– Group II Revenue = $200,000 (20,000/yr. X 10)

• With Complete Information– w = MRP– Group I wage = $10,000/yr.– Group II wage = $20,000/yr.

• With Asymmetric Information– w = average productivity– Group I & II wage = $15,000

223ME Slot 2

Signaling With Education to Reduce Asymmetric

Information• y = education index (years of higher

education)– C = cost of attaining educational level y– Group I--CI(y) = $40,000y

– Group II--CII(y) = $20,000y

– Assume education does not increase productivity

– Decision Rule:• y* signals GII and wage = $20,000

• Below y* signals GI and wage = $10,000

224ME Slot 2

Signaling

Years ofCollege

Value ofCollege

Educ.

0

$100K

Value ofCollege

Educ.

Years ofCollege

1 2 3 4 5 6 0 1 2 3 4 5 6

$200K

$100K

$200KCI(y) = $40,000y

Optimal choice of y for Group I

B(y) B(y)

y* y*

•Benefits = $100,000•Cost CI(y) = 40,000y• => Choose no education if•$100,000<$40,000y*• =>y* > 2.5

CII(y) = $20,000y

Optimal choice of y for Group I

•Benefits = $100,000•Cost CII(y)= 20,000y• => Choose y* if•$100,000<$20,000y*• =>y* < 5

225ME Slot 2

Cost/Benefit Comparison

• Decision rule works if y* is between 2.5 and 5– If y* = 4

• Group I would choose no school

• Group II would choose y*

• Rule discriminates correctly

• Education thus provides a useful signal about individual work habits.

226ME Slot 2

Role of Guarantees and Warranties

– Signaling to identify high quality and dependability

– Effective decision tool because the cost of warranties to low-quality producers is too high

– Further example of the ladder hypothesis of US historians

227ME Slot 2

Moral Hazard• Moral hazard occurs when the insured party’s

behavior changes after the fact to the detriment of the insurer.

• Determining the Premium for Fire Insurance– Warehouse worth $100,000– Probability of a fire:

• .005 with a $50 fire prevention program

• .01 without the program– With the program, the premium is:

• .005 x $100,000 = $500– Once insured owners purchase the insurance, the

owners no longer have an incentive to run the program, therefore the probability of loss is .01

– $500 premium will lead to a loss because the expected loss is not $1,000 (.01 x $100,000)

228ME Slot 2

Effects of Moral Hazard

Miles per Week0

$0.50

50 100 140

Costper

Mile

$1.00

$1.50

$2.00

D = MB (marginal benefit)

MC’

With moral hazard insurance companies cannot

measure mileage. MC to $1.00 andmiles driven increases to 140

miles/week--inefficient allocation.

MC

MC is the marginal costof driving. With no moral hazard

and assuming insurance companies can measure milesdriven MC = MB at $1.50 and

100 miles/week--efficient allocation.

229ME Slot 2

Main Features of the Principal--Agent Problem

• Agency Relationship – One person’s welfare depends on

what another person does

• Agent– Person who acts & gets pre-

contracted remuneration

• Principal– Person whom the action effects –

i.e., the residual risk-bearer230ME Slot 2

An Example of Principal--Agent Problem

• Company owners are principals.

• Workers and managers are agents.

• Owners do not have complete knowledge.

• Employees may pursue their own goals and reduce profits.

231ME Slot 2

Observations on the Principal--Agent Problem in Private

Enterprises

– Only 16 of 100 largest corporations have individual family or financial institution ownership exceeding 10%.

– Most large firms are controlled by management.– Monitoring management is costly (asymmetric

information).– Managers may pursue their own objectives.

• Growth• Utility from job

– Limitations to managers’ ability to deviate from objective of owners

• Stockholders can oust managers• Takeover attempts• Market for managers who maximize profits

232ME Slot 2

Observations on the Principal--Agent Problem in Public

Enterprises

• Managers’ goals may deviate from the principals’ goal (size)

• Monitoring is difficult (asymmetric information)

• Market forces are lacking

– Limitations to Management Power• Managers choose a public service position

• Managerial job market

• Legislative and agency oversight Competition among agencies

233ME Slot 2

Are non profit organizations more or less efficient than for-profit firms?

– 725 hospitals from 14 hospital chains

– Return on investment (ROI) and average cost (AC) measured

Return on Investment

19771981

For-Profit 11.6% 12.7%

Nonprofit 8.8% 7.4% 234ME Slot 2

Incentives in the Principal-Agent Framework

• Designing a reward system to align the principal and agent’s goals--an example

• Revenue also depends, in part, on luck and effort.

• High monitoring cost makes it difficult to assess the repair-person’s work

Bad luck Good luck

Low effort (a = 0) $10,000 $20,000

High effort (a = 1) $20,000 $40,000

235ME Slot 2

Example of a reward system to align the principal and agent’s

goals• Owners cannot determine a high or low effort when revenue = $20,000• Repairperson’s goal is to maximize wage net of cost• Cost = 0 for low effort• Cost = $10,000 for high effort = his opportunity cost under a=0• w(R) = repairperson wage based only on output

– Choosing a wage-bonus scheme to induce high effort:• If R=10,000 or 20,000 then w=0• And if R=40,000, w=24,000• Then for a=0, w=0

• For a=1, w=(24,000+0)/2=12,000, i.e., Net wage = $2,000• Owner’s expected revenue=30,000 and expected profit=(20,000

+40,000-24,000)/2=18,000– Choosing an alternative revenue-sharing arrangement to induce

high effort:• w = R - $18,000 if R>=18,000 and w=0 if R<18,000• Then for a=0, expected w=(0+20,000-18,000)/2=1,000• For a=1, expected w=(20,000-18,000 +40,000-18,000)/2=12,000, i.e., net

gain=2,000 => Owner’s expected net profit = 18,000 as before– Conclusion : Incentive structure that rewards the outcome of

high levels of effort can induce agents to aim for the goals set by the principals.

236ME Slot 2

Without shirking, the market wageis w*, and full-employment exists at L*

Demand forLabor

w*

L*

SL

Unemployment in a Shirking Model

Quantity of Labor

WageNo-ShirkingConstraint

The no-shirkingconstraint gives

the wage necessaryto keep workers

from shirking.

we

Le

At the equilibrium wage, We the firm hires Le workers,creating unemployment of L* - Le.

237ME Slot 2

Efficiency Wages at Ford Motor Company

• Labor turnover at Ford – 1913: 380%– 1914: 1000%

• Initial average pay = $2 - $3• Ford increased pay to $5

• Results– Productivity increased 51%– Absenteeism was halved– Profitability rose from $30 million in

1914 to $60 million in 1916.

238ME Slot 2

An Algebraic Expression for Efficiency Wages

• g = p (we – w*) N, whereg = marginal gain from shirking,we = efficiency wage ratew* = market-clearing wage rateN = time-horizon of the laborer=> we = w* + g / (pN)=> we > w* and further=> we = f( W*, g, p, N)

(+) (+) (-) (-)

239ME Slot 2

ExternalitiesExternalitiesExternalitiesExternalitiesME Session 19, August 31, ME Session 19, August 31,

20102010

Prof. Samar K. DattaProf. Samar K. Datta

Overview• Externalities

– Negative: Action by one party imposes a cost on another party

– Positive: Action by one party benefits another party

• Ways of Correcting Market Failure

– Recycling

241ME Slot 2

External Cost• Scenario

– Steel plant dumping waste in a river– The entire steel market effluent can be

reduced by lowering output (fixed proportions production function)

– Marginal External Cost (MEC) is the cost imposed on fishermen downstream for each level of production.

– Marginal Social Cost (MSC) is MC plus MEC.

242ME Slot 2

MC

S = MCI

D

P1

Aggregate social cost of

negativeexternality

P1

q1 Q1

MSC

MSCI

When there are negativeexternalities, the marginalsocial cost MSC is higher

than the marginal cost.

External Costs

Firm output

Price

Industry output

Price

MEC

MECI

The differences isthe marginal external

cost MEC.

q*

P*

Q*

The industry competitiveoutput is Q1 while the efficient

level is Q*.

The profit maximizing firmproduces at q1 while the

efficient output level is q*.

243ME Slot 2

Externalities

• Positive Externalities and Inefficiency– Externalities can also result in too

little production, as can be shown in an example of home repair and landscaping.

• Negative Externalities encourage inefficient firms to remain in the industry and create excessive production in the long run.

244ME Slot 2

MCP1

External Benefits

Repair Level

Value

D

Is research and development discouraged by positive

externalities?

q1

MSB

MEB

When there are positiveexternalities (the benefitsof repairs to neighbors),marginal social benefits

MSB are higher thanmarginal benefits D.

q*

P*

A self-interested home ownerinvests q1 in repairs. Theefficient level of repairs

q* is higher. The higher priceP1 discourages repair.

245ME Slot 2

Ways of Correcting Market Failure in

Pollution

• Assumption: Fixed-proportion production technology

• Must reduce output to reduce emissions• Use an output tax to reduce output

• Input substitution possible by altering technology

• Options for Reducing Emissions– Emission Standard

• Set a legal limit on emissions at E* (12)• Enforced by monetary and criminal penalties• Increases the cost of production and the

threshold price to enter the industry– Emissions Fee

• Charge levied on each unit of emission246ME Slot 2

The Efficient Level of Emissions

Level of Emissions

2

4

6

Dollarsper unit

of Emissions

0 2 4 6 8 10 12 14 16 18 20 22 24 26

MSC

MCAE*

The efficient level ofemissions is 12 (E*) where

MCA = MSC.

At Eo the marginalcost of abating emissions

is greater than themarginal social cost.

E0

At E1 the marginalsocial cost is greater

than the marginal cost of abatement.

E1

Assume:1) Competitive market2) Output and emissions decisions are independent3) Profit maximizing output chosen

Why is this more efficient than zero emissions?

247ME Slot 2

TotalAbatement Cost

Cost is less than thefee if emissions were

not reduced.

Total Feeof Abatement

Standards and Fees

Level of Emissions

Dollarsper unit

of Emissions MSC

MCA

3

12E*

Fee

Level of abatement

248ME Slot 2

Firm 2’s ReducedAbatement

Costs

Firm 1’s IncreasedAbatement Costs

MCA1

MCA2

The Case for Fees

Level of Emissions

2

4

6

Fee perUnit of

Emissions

0 1 2 3 4 5 6 7 8 9 10 11 12 13

1

3

5

14

The cost minimizing solutionwould be an abatement of 6

for firm 1 and 8 for firm 2 andMCA1= MCA2 = $3.

3.75

2.50

The impact of a standard ofabatement of 7 for both firms

is illustrated.Not efficient because

MCA2 < MCA1.

If a fee of $3 was imposedFirm 1 emissions would fall

From 14 to 8. Firm 2 emissionswould fall from 14 to 6.

MCA1 = MCA2: efficient solution.

249ME Slot 2

• Standards Versus Fees– Assumptions

• Policymakers have asymmetric information

• Administrative costs require the same fee or standard for all firms

• Advantages of Fees– When equal standards must be used,

fees achieve the same emission abatement at lower cost.

– Fees create an incentive to install equipment that would reduce emissions further.

Ways of Correcting Market Failure

250ME Slot 2

ABC is the increasein social cost less thedecrease in abatement

cost.

MarginalSocialCost

Marginal Costof Abatement

The Case for Standards

Level of Emissions

Fee perUnit of

Emissions

0 2 4 6 8 10 12 14 16

2

4

6

8

10

12

14

16

E

Based on incompleteinformation standard is 9

(12.5% decrease).ADE < ABC

DA

B

C Based on incompleteinformation fee is $7

(12.5% reduction).Emission increases to 11.

251ME Slot 2

• Permits help develop a competitive market for externalities.

• Agency determines the level of emissions and number of permits

• Permits are marketable

• High cost firm will purchase permits from low cost firms

• Cost of Reducing (Sulfur Dioxide ) Emissions:– Conversion to natural gas from coal and oil– Emission control equipment

• Benefits of Reducing Emissions:– Health– Reduction in corrosion– Aesthetic

Transferable Emissions Permits as a Way of Correcting Market

Failure

252ME Slot 2

Sulfur Dioxide Emissions Reductions

Sulfur dioxide concentration (ppm)

20

40

60

0

Dollarsper

unit ofreduction

0.02 0.04 0.06 0.08

Marginal Social Cost

Marginal Abatement Cost

ObservationsObservations•MAC = MSC @ .0275MAC = MSC @ .0275•.0275 is slightly below actual emission level.0275 is slightly below actual emission level•Economic efficiency improvedEconomic efficiency improved

253ME Slot 2

Emissions Trading and Clean Air

• Bubbles– Firm can adjust pollution controls for

individual sources of pollutants as long as a total pollutant limit is not exceeded.

• Offsets– New emissions must be offset by

reducing existing emissions • 2000 offsets since 1979

254ME Slot 2

• Cost of achieving an 85% reduction in hydrocarbon emissions for DuPont– Three Options

• 85% reduction at each source plant (total cost = $105.7 million)

• 85% reduction at each plant with internal trading (total cost = $42.6 million)

• 85% reduction at all plants with internal and external trading (total cost = $14.6 million)

• 1990 Clean Air Act– Since 1990, the cost of the permits has

fallen from an expected $300 to below $100.

• Causes of the drop in permit prices – More efficient abatement techniques– Price of low sulfur coal has fallen

Emissions Trading and Clean Air

255ME Slot 2

The Efficient Amount of Recycling

Scrap

Cost

0 4 8 12

MCR

MSC

m*

With a refundable deposit,MC increases andMC = MSC = MCR.

MC + per unit refundMC

m1

Without market interventionthe level of scrap will be at m1

and m1 > m*.Households can dispose of glass and other garbage at very low cost. The low cost of disposal creates a divergence between the private and the social cost of disposal. S0, raising MC is a move in the right direction.

Extent of recycling256ME Slot 2

Refundable Deposits

Amount of Glass

$

D

Price falls to P’ and the amount of recycled glass increases to M*.

Sv

Sr

S

The supply of glass is the horizontal sum of the Supply of virgin glass (Sr) and the supply of recycled

glass (Sr).

M1

P

Without refunds the price of glass is P and

Sr is M1.

S’r

S’

P’

M*

With refunds Sr increasesto S’r and S increases to S’.

257ME Slot 2

Property Rights Property Rights and Public Goodsand Public GoodsProperty Rights Property Rights

and Public Goodsand Public GoodsME Session 20: Sept.6, 2010ME Session 20: Sept.6, 2010

Prof. Samar K. DattaProf. Samar K. Datta

Overview

• Property Rights & Coase Theorem

• Common Property Resources

• Public Goods

• Private Preferences for Public Goods

259ME Slot 2

Externality & Property Rights

• Property Rights– Legal rules describing what people or

firms may do with their property– For example

• If residents downstream owned the river (clean water) they control upstream emissions.

• Bargaining and Economic Efficiency– Economic efficiency can be achieved

without government intervention when the externality affects relatively few parties and when property rights are well specified.

260ME Slot 2

PRIVATE SOLUTION TO EXTERNALITIES

• Coase Thorem: Economic agents can arrive at an efficient solution (i.e., an optimum assignment of property rights)– irrespective of initial assignment of property rights, – provided they can bargain free of cost (i.e., w/o

transaction costs), and – there is no wealth effect to thwart the bargaining

process.

• Coase Theorem at Work: Negotiating an Efficient Solution - 1987 --- New York garbage spill (200 tons) littered the New Jersey beaches– The potential cost of litigation resulted in a solution

that was mutually beneficial to both parties.

261ME Slot 2

Coase Theorem

MC of pollution to MC of pollution to fisheriesfisheries

MC of pollution MC of pollution abatement by factoryabatement by factory

PollutionPollution Abatement Abatement

AA

CC

BB

EE O’O’OO FF

LL

XX

OE is optimal quantity of pollution and O’E is the corresponding optimal quantity of abatement

At E, the marginal costs of pollution and abatement are equal, and the sum of the total costs i.e. triangle OXO’ is the least

At A, MC of abatement exceeds MC of pollution, so it is cheaper to compensate the fishermen AB than to abate the pollution AC

Demand price for abatement falls short

Demand price forabatement now higher to facilitate pollution reduction

262ME Slot 2

Common Property Resources

• Common Property Resource– Everyone has free access.– Likely to be over-utilized– Examples

• Air and water• Fish and animal populations• Minerals

• Solution– Private ownership

• Question– Wouldn’t private ownership be

impractical?

263ME Slot 2

Common Property Resources

Fish per Month

Benefits,Costs($ per

fish)

Demand

However, private costsunderestimate true cost.

The efficient level of fish/month is F* where

MSC = MB (D)

Marginal Social Cost

F*

Private Cost

FC

Without control the numberof fish/month is FC where

PC = MB (marginal benefit).

264ME Slot 2

Crawfish Fishing in Lousiana

• Finding the Efficient Crawfish Catch– F = crawfish catch in millions of pounds/yr– C = cost in dollars/pound

• Demand– C = 0.401 = 0.0064F

• MSC– C = -5.645 + 0.6509F

• PC– C = -0.357 + 0.0573F

• Efficient Catch– 9.2 million pounds– D = MSC

265ME Slot 2

Crawfish Catch(millions of pounds)

CCost

(dollars/pound)

Demand

Marginal Social Cost

Private Cost

Crawfish as a CommonProperty Resource

11.9

2.10

9.2

0.325

266ME Slot 2

Public Goods• Public Good Characteristics

– Non-rival• For any given level of production the marginal

cost of providing it to an additional consumer is zero.

– Non-exclusive• People cannot be excluded from consuming

the good.

• Not all government produced goods are public goods– Some are rival and non-exclusive (more like

a common property resource)

• Education 267ME Slot 2

Typology of Goods

Characteris-tics

ExcludableNon-

excludable

Rival Private GoodCommon Property Resource

Non-rival Club Good Public Good

268ME Slot 2

D1

D2

D

When a good is non-rival, the social marginalbenefit of consumption (D) , is determined by

vertically summing the individual demand curves for the good.

Efficient Public Good Provision

Output0

Benefits(dollars)

1 2 3 4 5 6 7 8 109

$4.00

$5.50

$7.00

Marginal Cost

$1.50

Efficient output occurswhere MC = MB at 2

units of output. MB is$1.50 + $4.00 or $5.50.

Horizontal sum of demand curves

What if this is the MC curve?

269ME Slot 2

Problem with Public Goods

• Free Riders– There is no way to provide some goods

and services without benefiting everyone.– Households do not have the incentive to

pay what the item is worth to them.– Free riders understate the value of a good

or service so that they can enjoy its benefit without paying for it.

• Clean Air is a public good– Non-exclusive and non-rival

• What is the price of clean air?270ME Slot 2

How to know Private Preferences for Public

Goods?

• Government production of a public good is advantageous because the government can assess taxes or fees to pay for it.

• Determining how much of a public good to provide when free riders exist is however extremely difficult.

271ME Slot 2

The Demand for Clean Air

Nitrogen Oxides (pphm)0

Dollars

1 2 3 4 5 6 7 8 109

2000

2500

3000

500

1500

1000

Low Income

Middle Income

High Income

272ME Slot 2

Findings on Demand for Clean Air

– Amount people are willing to pay for clean air increases substantially as pollution increases.

– Higher income earners are willing to pay more (the gap between the demand curves widen)

– National Academy of Sciences found that a 10% reduction in auto emissions yielded a benefit of $2 billion---somewhat greater

than the cost.

273ME Slot 2