Perfect Competition Many buyers and sellers Homogeneous product Perfect information No transaction costs Free entry and exit

Key Implications Firms are “price takers” (P = MR) In the short-run, firms may earn profits or losses Long-run profits are zero

Unrealistic? Why Learn? Many small businesses are “price-takers,” and decision rules for such firms are similar to those

of perfectly competitive firms It is a useful benchmark Explains why governments oppose monopolies Illuminates the “danger” to managers of competitive environments

Importance of product differentiation Sustainable advantage

Setting Price

Setting Output: MR = MC MR = P, therefore Set P = MC to maximize profits

A Numerical Example Given

P=$10 C(Q) = 5 + Q2

Optimal Price? P=$10

Optimal Output? MR = P = $10 and MC = 2Q 10 = 2Q Q = 5 units

Maximum Profits? PQ - C(Q) = (10)(5) - (5 + 25) = $20

Long Run Adjustments? If firms are price takers but there are barriers to entry, profits will persist If the industry is perfectly competitive, firms are not only price takers but there is free entry

Other firms enter the marketEffect of Entry on Price?

Effect of Entry on the Firm’s Output and Profits?

Summary of Logic Short run profits leads to entry Entry increases market supply, drives down the market price, increases the market quantity

Demand for individual firm’s product shifts down Firm reduces output to maximize profit Long run profits are zero

Features of Long Run Competitive Equilibrium P = MC

Socially efficient output P = minimum AC

Efficient plant size Zero profits

Firms are earning just enough to offset their opportunity cost

Monopoly Single firm serves the “relevant market” Most monopolies are “local” monopolies The demand for the firm’s product is the market demand curve Firm has control over price

But the price charged affects the quantity demanded of the monopolist’s product“Natural” Sources of Monopoly Power

Economies of scale Economies of scope

“Created” Sources of Monopoly Power Patents and other legal barriers (like licenses) Tying contracts Exclusive contracts Collusion

A Monopolist’s Marginal Revenue

Monopoly Profit MaximizationProduce where MR = MC.Charge the price on the demand curve that corresponds to that quantity.

A Useful Formula What’s the MR if a firm faces a linear demand curve for its product? P(Q) = a + bQ

MR = a + 2bQA Numerical Example

Given estimates of P = 10 - QC(Q) = 6 + 2Q

Optimal output? MR = 10 - 2QMC = 2

10 - 2Q = 2Q = 4 units

Optimal price? P = 10 - (4) = $6

Maximum profits? PQ - C(Q) = (6)(4) - (6 + 8) = $10

Long Run Adjustments? None, unless the source of monopoly power is eliminated.

Why Government Dislikes Monopoly? P > MC

Too little output at a too high price Deadweight loss of monopoly

Deadweight Loss of Monopoly

Arguments for Monopoly The beneficial effects of economies of scale, economies of scope on price and output may

outweigh the negative effects of market power Encourages innovation

Oligopoly Relatively few firms, usually less than 10.

Duopoly - two firms The products firms offer can be either differentiated or homogeneous.

Role of Strategic Interaction What you do affects the profits of your rivals What your rival does affects your profits

Cournot Model A few firms produce goods that are either perfect substitutes (homogeneous) or imperfect

substitutes (differentiated) Firms set output, as opposed to price Each firm believes their rivals will hold output constant if it changes its own output (The output

of rivals is viewed as given or “fixed”) Barriers to entry exist

Reaction Functions Suppose two firms produce homogeneous products. Firm 1’s reaction (or best-response) function is a schedule summarizing the amount of Q1 firm

1 should produce in order to maximize its profits for each quantity of Q2 produced by firm 2. Since the products are substitutes, an increase in firm 2’s output leads to a decrease in the

profit-maximizing amount of firm 1’s product.Graphically

Cournot Equilibrium Situation where each firm produces the output that maximizes its profits, given the the output

of rival firms No firm can gain by unilaterally changing its own output

Cournot Equilibrium

Summary of Cournot Equilibrium The output Q1

* maximizes firm 1’s profits, given that firm 2 produces Q2*

The output Q2* maximizes firm 2’s profits, given that firm 1 produces Q1

* Neither firm has an incentive to change its output, given the output of the rival Beliefs are consistent:

In equilibrium, each firm “thinks” rivals will stick to their current output -- and they do!Firm 1’s Isoprofit Curve

The combinations of outputs of the two firms that yield firm 1 the same level of profit

Another Look at Cournot Decisions:

Another Look at Cournot Equilibrium

Stackelberg Model Few firms

Producing differentiated or homogeneous products Barriers to entry Firm 1 is the leader

The leader commits to an output before all other firms Remaining firms are followers.

They choose their outputs so as to maximize profits, given the leader’s output.Equilibrium

Summary Stackelberg model illustrates how commitment can enhance profits in strategic environments Leader produces more than the Cournot equilibrium output

Larger market share, higher profits First-mover advantage

Follower produces less than the Cournot equilibrium output Smaller market share, lower profits

Bertrand Model Few firms

Firms produce identical products at constant marginal cost. Each firm independently sets its price in order to maximize profits

Barriers to entry Consumers enjoy

Perfect information Zero transaction costs

Equilibrium

Firms set P1 = P2 = MC! Why? Suppose MC < P1 < P2 Firm 1 earns (P1 - MC) on each unit sold, while firm 2 earns nothing Firm 2 has an incentive to slightly undercut firm 1’s price to capture the entire market Firm 1 then has an incentive to undercut firm 2’s price. This undercutting continues... Equilibrium: Each firm charges P1 = P2 =MC When firms have the same cost structure, price competition drives prices to the unit costs level

and firms make zero profit. Therefore, the industry equilibrium is independent of the firms number: P=Cmg even if there

are more 2 than competitors. Paradox

Even if there are only 2 competitors, prices will be set at the level of marginal cost (the perfect competition outcome).

However, many industries look like Bertrand model but where prices are higher than marginal cost. This outcome might be due to:

Capacity constraints Product Differentiation Dynamic interaction

Contestable Markets Key Assumptions

Producers have access to same technology Consumers respond quickly to price changes Existing firms cannot respond quickly to entry by lowering price Absence of sunk costs

Key Implications Threat of entry disciplines firms already in the market Incumbents have no market power, even if there is only a single incumbent (a

monopolist)Summary

Different oligopoly scenarios give rise to different optimal strategies and different outcomes Your optimal price and output depends on …

Beliefs about the reactions of rivals Your choice variable (P or Q) and the nature of the product market (differentiated or

homogeneous products) Your ability to commit

Collusion - Cartels and Tacit Collusion

Dynamic Price Competition Consider a market with n firms producing products that are perfect substitutes. Current market price P0

Each firm has constant marginal cost c. In each period market demand is D(P) If all firms set the same price, demand is equally divided among them. PM is monopoly price (price level that maximizes industry profits) Industry profit in each period: P(P)=(P-c)D(P) We assume that P0< PM Maximum industry profit: Monopoly profit PM=(PM-c)D(PM) Profit level when the market price is P0: P0 We assume : PM/n< P0

Firms simultaneously set prices in each period of time. If all firms, but one (firm D), raise their prices to P M, the firm that does not follow that strategy

and sticks with the current price (keeps the entire market) makes higher profit P0 — Prisoners’ Dilemma situation

Is it possible to get the monopoly price as the outcome of a noncooperative game, without collusion among firms?

Consider that all firms play the following grim strategy: In the first period each firm raises its price to PM, the monopoly level. In each subsequent period, each firm observes past prices before setting its own price: If all firms have respected PM than each firm will set PM. If a firm observes its rival setting a lower price it punishes the deviation by reverting to

Po.Firm D deviates

Suppose that firm D sticks with current price Pº whereas all other firms increase prices to PM. Firm D assumes that all firms play according to the grim strategy.

Prices evolution when Firm D deviatesPeriod 1 2 3…..Firm D P0 P

0P0

Other Firms PM

P0

P0

Profit evolution for Firm DPeriod 1 2 3Deviates Po P0/n P0/nCooperates PM/n PM/n PM/n

Discounted present value of firm D’s profit Assume that i is the discount rate Firm D deviates:

Firm D cooperates

Firm D will cooperate if :

Each firm must be sufficiently “patient” Coordination and equilibrium

In an infinite repeated game a collusive outcome can emerge as the equilibrium in an oligopoly. In order to achieve this outcome firms must solve a coordination problem. However, coordination without explicit agreements or communication is very difficult.

Solution for the Coordination Problem In order to have a collusive outcome we must have a strategy (like the grim strategy) that

works as a Focal Point: Each firm expects that all rivals will follow that strategy.

Factors that facilitate Collusion Collusive pricing is more likely to be an equilibrium the greater the frequency with which firms

interact and the greater the probability of continuation and growth of industry. Collusion is normally easier to maintain among few and similar firms. Collusion is usually easier to maintain when firms compete in more than one market.

We are in New York City and we focus on the stereo price wars. One of the main competitors, Crazie Eddie has made his trademark “We cannot be undersold. We will not be undersold. Our prices are the lowest – guaranteed. Our prices are insane.” His main competitor, Newmark & Lewis, is no less ambitious. With any purchase, you get the store’s “Lifetime low-price guarantee.” It promises to rebate double the difference if you can find a lower price elsewhere. “If, after your purchase, you find the same model advertised or available for sale for less (confirmed printed proof required) by any other local stocking merchant, in this marketing area, during the lifetime of your purchase, we, Newmark and Lewis, will gladly refund (by check) 100% of the difference, plus an additional 25% of the difference or, if you prefer, Newmark and Lewis will give you a 200% gift certificate refund (100% of the difference plus an additional 100% of the difference, in gift certificates).”from Newmark and Lewis’s Lifetime Low-Price Gurantee»

Market Structure and Market Power

Market Definition Market Structure: refers to the number and characteristics of firms in a market. Market: set of suppliers producing the same product or a set of products that are close substitutes.

high cross-price elasticity of demand

How can we define those set of products?

1n

[ΠM−Π0 ]

Π0−1nΠM

≥i

The set of products that define a market must have very high cross price-demand elasticities among them and very low cross price-demand elasticities vis à vis other products.

Price Elasticities of Demand for Different Car Models

Nissan Sentra Ford Escort Toyota Lexus BMW 745i

Nissan Sentra -6.5282 0.4544 0.0008 0.0000

Ford Escort 0.0778 -6.0309 0.0008 0.0000

Toyota Lexus 0.0002 0.0010 -3.0847 0.0322

BMW 745i 0.0001 0.0005 0.0926 -3.5151

Close substitutes Products are close substitutes if:

Products have identical or similar nature and characteristics; Products are sold in the same geographic market;

Relevant Market The relevant market for one product includes all products that significantly limit the price of this product. The extent to which firms are able to increase their prices above normal competition levels depends on the possibility

for consumers to buy substitute goods. The fewer the substitute products the less elastic the demand curve is and the more probable is to find higher prices.

Ex: plastic bottles and glass bottles

Why is it so difficult to define a Market? Frequently product differentiation is due to small characteristics of the product.

Ex: Diet Coke belongs to Light cola market Cola market Soft drinks market

Relevant Geographic Market: Imports Transportation Costs

Characterization of Market Structure Concentration in the market

Number of firms Size of firms

Measuring Market Structure 1. k-firm concentration ratio

Combined share of the k largest firms in the market

2. Herfindahl Index:

H index equals the sum of the squared market shares of all firms in the market Conveys more information than the Ck ratio:

Sensitive to changes in the sizes of the largest firms in the market.

Concentration Index

C k≡∑i=1

k

s i where si is the market share of firm i

Ck∈[ kn ,1]H≡∑

i=1

n

si2 H∈[ 1

n,1]

To calculate H index it is usually sufficient to consider

In the US the Herfindahl index is used to determine whether mergers should be examined; increases of over 100 points generally provoke scrutiny, although this varies from case to case. The Antitrust Division of Department of Justice considers Herfindahl indices between 1000 and 1800 to be moderately concentrated and indices above 1800 to be concentrated. As the market concentration increases, competition and allocative efficiency decrease and collusion and market power increase.

3. Numbers-equivalent of firms (Adelman): The Herfindahl index in a market with N identical firms is 1/N. Because of this property the reciprocal of the H index is referred to as the numbers–equivalent of firms.

NE=1/HEx: H=.125 NE=8

The Herfindahl Index can be expressed as a function of: Total number of firms in the market Variance of the distribution of the firms’ market shares

If all firms have equal (identical) shares (that is, if the market structure is completely symmetric, in which case si = 1/n for all i) then σ2 is zero and H equals 1/n. If the number of firms in the market is held constant, then a higher variance due to a higher level of asymmetry between firms' shares (that is, higher share dispersion) will result in a higher value of the H index.

Ck and Herfindahl

An example:

Market Structure and Market Power Lerner Index. Is a measure of market power and is defined as the weighted average of each firm’s margin, with weight

given by firms’ market shares.

Bain Paradigm

Structure® Conduct ® Performance1. Structure:

Suppliers and buyers: number and size Product Differentiation Entry conditions

Minimum efficient scale Sunk costs

2. Conduct Collusion Strategic Behaviour

• Pricing (predatory)• Advertising and R&D (Product Differentiation)• Vertical Integration, Vertical Integration

H=nσ2+ 1n

3. Performance Profitability, efficiency

Topic 3 - Price DiscriminationOverviewI. Extracting Consumer Surplus

Price Discrimination Two-Part Pricing Block Pricing Commodity Bundling

II. Pricing for Special Cost and Demand Structures Peak-Load Pricing Cross Subsidies

Pricing The most difficult marketing decision for managers Which is the main goal? Which is the right price? Through Pricing the firm tries to capture most of the value created by its own product. But the costs ...

Key factors Demand elasticity Costs Complements and Substitutes Dynamics: loyalty, experience curve Information Psychological aspects Competition

Standard Pricing and Profits

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

A Simple Markup Rule Suppose the elasticity of demand for the firm’s product is EF 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 demand, lower markup. Less elastic demand, higher markup.

An Example Elasticity of demand for Dell computers is -2 P = [EF/(1+ EF)] ´ MC P = [-2/(1 - 2)] ´ MC P = 2 ´ MC Price is twice marginal cost Fifty percent of Dell’s price is margin above manufacturing costs.

Markup as a pricing strategy Frequently firms set prices equal to a markup over average total costs.

o P=ATC(1+mk) or markup=(P-ATC)/ATC

Problems:o Ignores market conditionso Ignores demand volatilityo Ignores competitors pricing strategieso Ignores that demand is a function of priceo Uses ATC instead of MgC

Benefits:o It is easy to use.

Determinants of Demand Elasticity Who’s paying? Is it the decision maker? Does the product cost represent a significant % of total expense? Is the buyer the final consumer? If not, does he compete in prices in the product market? In this market do high prices signal high quality? Are search costs relevant to consumer? Is timing relevant for consumer? Can consumers compare prices and performances of different alternatives? Are switching costs relevant to consumer?

Market Segmentation A pricing strategy depends upon:

o Costs o Price Sensitivity o Competition

If any of these factors varies significantly across different market segments it is worthwhile to use sophisticated marketing

Price Discrimination

Railroad Tariffs for Passengers Dupuit (1849) “It is not because of the few thousand francs which would have to be spent to put a roof over the third-class

carriages or to uphoslter the third-class seats that some company or other has open carriages with wooden benches... What the company is trying to do is to prevent the passengers who can pay the second class fare from traveling third-class; it hits the poor, not because it wants to hurt them, but to frighten the rich.”

Price discrimination Price discrimination is more frequent on services sales than on goods sales

– Services are usually consumed in the purchasing moment; – Services are frequently “perishable goods”, they are not potentially storable in order to be sold later.

Price Discrimination Under price discrimination we design price schemes that are tailor made for different consumer segments. We must have:

– Separation among consumer segments– Each consumer segment has a different price – Segments with different price elasticity

First-Degree or Perfect Price Discrimination Practice of charging each consumer the maximum amount he or she is willing to pay for each incremental unit Allows a firm to extract all surplus from consumers

Perfect Price Discrimination

Caveats: In practice, transactions costs and information constraints make this rather difficult to implement perfectly (but

car dealers and some professionals come close). Price discrimination won’t work if consumers can resell the good.

Second Degree Price Discrimination The practice of posting a discrete schedule of declining prices for different quantities. Example: Electric utilities

Second Degree Price Discrimination

Second Degree Price Discrimination

Quantity discount

Third Degree Price Discrimination The practice of charging different groups of consumers different prices for the same product Examples include student discounts, senior citizen’s discounts, regional & international pricing

Implementing Third Degree Price Discrimination Suppose the total demand for a product is comprised of two groups with different elasticities, E1 < E2

Notice that group 1 is more price sensitive than group 2 Profit-maximizing prices? P1 = [E1/(1+ E1)] ´ MC P2 = [E2/(1+ E2)] ´ MC

An Example Suppose the elasticity of demand for Dell computers in the US is EU = -1.5, and the elasticity of demand in Japan is

EJ = -2.5 Marginal cost of manufacturing a computer 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 Dell’s optimal third-degree pricing strategy is to charge a higher price in the US, where demand is less elastic

Case Studies Coupons: why do firms use coupons instead of reducing prices? Hotels Pricing policy Pricing for flight tickets

Price discrimination Restaurants lunch/dinner

– Big Mac is an alternative “Early Bird” Special

– Parking, breakfast price Intertemporal Price Discrimination

Peak-Load Pricing If during peak times the demand is higher than the capacity of the firm, the firm should engage in peak-load

pricing – Goal: try to smooth demand

Occurs when demand fluctuates along a time period (day, week, month) but supply is not “storable”– Ex: plains, hotels, electricity, telephones

Since you cannot adjust supply throughout time you will try to manage demand– Charge a higher price (PH) during peak times (DH) – Charge a lower price (PL) during off-peak times (DL)

Peak Load Pricing

Two-Part Pricing When it isn’t feasible to charge different prices for different units sold, but demand information is available, 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 membershipsHow Two-Part Pricing Works

1. Set price at marginal cost.2. Compute consumer surplus.3. Charge a fixed-fee equal to consumer surplus.

Heterogeneous Consumers

Heterogeneous Consumers

Heterogeneous Consumers

2 different groups of consumers Separation: P=MgC and

– TL=s1+s2+s3– TH=s1+s2+s3+s4+s5+s6

Uniform Tariff: trade-off between a too high tariff (light users will exit the market) and a too low tariff (it will be adopted by heavy users)

– 1ª hip.: T=s1+s2+s3 e P=MgC; H users gain surplus=s4+s5+s6 and Profit=2*T– 2ª Hip: T=s1 e P>MgC; Profit=2*(s1+s2)+s3+s4– Firm’s profits increase iff s4>s3.

Block Pricing The practice of packaging multiple units of a product together and selling them as one package. Examples

– Paper– Six-packs of drinks

An Algebraic Example Typical consumer’s demand is P = 10 - 2Q C(Q) = 2Q Optimal number of units in a package? Optimal package price?

Optimal Quantity to Package: 4 Units

Optimal Price for the Package: $24

Costs and Profits with Block Pricing

Commodity Bundling The practice of bundling two or more products together and charging one price for the bundle Examples

– Vacation packages– Computers and software– Perfume and body cream

An Example

Film A Film B

Theater X $12 000 $3 000

Theater Y $10 000 $4 000

Solution: Separated Prices: PA=$10 000; PB=$3 000;

TR= 2x($10 000+$3 000)= $26 000 Bundle: Pbundle=$14 000; TR=2x$14 000=$28 000

An example

Film A Film B

Theater X $12 000 $4 000

Theater Y $10 000 $3 000

Solution: Separated prices: PA=$10 000; PB=$3 000;

TR= 2x($10 000+$3 000)= $26 000 Bundle: Pbundle=$13 000; TR=2x$13 000=$26 000

Mixed Bundling vs. Pure Bundling

Recap of Pricing Strategies First degree price discrimination, block pricing, and two part pricing permit a firm to extract all consumer surplus. Commodity bundling, second-degree and third degree price discrimination permit a firm to extract some (but not all)

consumer surplus. Simple markup rules are the easiest to implement, but leave consumers with the most surplus and may result in

double-marginalization.

Different strategies require different information.Topic 4: Product Differentiation

Product differentiation Competing firms differentiate their products in such way that products fulfilling the same need typically do not

have identical features. The differentiation of goods along key features and minor details is an important strategy for firms to defend from

intensive price competition and gain market power.Horizontal Differentiation (variety)

Horizontal differentiation occurs in a market when products are different according to features that make them more attractive for certain groups of customers and less attractive for others groups of customers.

Those features cannot be ordered. Ex: considering ice-cream tastes, chocolate is not “better” than lemon. Red cars are not “better” than blue cars. With the same price, different consumers buy different products.

Vertical Differentiation (quality) Vertical differentiation emerges in a market where products can be ordered according to their objective quality.

Ex: Mercedes and Fiat Firms that succeed in increasing perceived quality of their goods can set higher prices because some consumers

are willing to pay more for additional quality. If prices were identical all consumers would prefer to buy the product with higher quality.

Horizontal Product Differentiation: Hotelling ModelModelo de Hotelling

Hotelling (1929) criticized Bertrand because he assumed unrealistic product homogeneity.o A slightly decrease in one firm’s price allows it to capture the entire market.o Demand is discontinuous

In the real world, a small decrease in one firm’s price usually only deviates a small number of the competitors’ customers.

Modelo de Hotelling Introduces differentiation in one dimension: the distance between the consumer location and the firm location. Geographic location is a differentiation factor. Distance implies the existence of transportation costs:

o Direct Cost: bus ticket, gas costo Indirect Cost: opportunity cost of time

Market Structure Assume that the market is a straight line with fixed length 1

o Ex: the main street of one city or a beach Identical consumers are uniformly distributed along the line There are two firms in the market (A e B ) that sell an identical product: Both firms have constant and identical average and marginal cost that we assume equal to zero.

Market structure Firm A is located at a distance a, from the left end of the market; Firm B is located at a distance b from the right

end of the market.

Market Structure Firms can relocate without costs. t is the transportation cost per distance unit Each consumer pays the firm’s price P and the transportation cost

o PA + t x Price paid to firm A by a consumer located at a x distance from firm A. There is a consumer in each point of the line and each consumer always buys one unit of the good. Each consumer will choose either firm as long as the combined price P and transportation cost of the product is

less than the competitive firm.Location : x* Indifferent Consumer

X* is indifferent between firm A or firm B

Indifferent Consumer Location: x*

Firm A’s demand consists in all consumers located to left of point x*.

When a ≤ x* ≤ 1-b, x* is the frontier between consumers that buy from firm A and from firm B.Another case

Now firm A captures the entire market.

Another different case And now firm A keeps only its captive market

The market is served by the two firms iff x* is located between a and 1-b

Condition for a<x*<1-b X*= (PB-PA)/2t+(1-b+a)/2 x>a : PA < PB+t(1-b-a) x<1-b: PA> PB-t(1-b-a) a<x<1-b se PB-t (1-b-a) < PA < PB+t(1-b-a)

Demand for firm A qA=0 if PA >PB+t(1-b-a) qA=(PB-PA)/2t+(1-b+a)/2 if PB +t(1-b-a)≥ PA ≥ PB -t(1-b-a) qA= 1 if PA <PB-t(1-b-a)

Profit Function for Firm A

Nash equilibrium for a price game Is a pair (PA*, PB*) such that PA* is the best response against PB* and vice-versa. Suppose that a=1-b, that is, assume that both firms are located on the market center.

In this case the model is equivalent to the Bertrand model and there exists a unique equilibrium given by: PA*= PB*= c =0

There is no differentiation.…Assume that prices are fixed by a regulator

In this case prices are identical and independent to firms’ locations.o PA= PB =P

Which will be the equilibrium location for each firm? Indifferent consumer: x*=(1-b+a)/2

Which is the Nash Equilibrium? If firm B is located to the right of the central point (1-b>1/2) then the firm A’s best response is to relocate to the

left of firm B’s location. But then the best response of firm B is to relocate to the left of firm A’s location and...

The Nash equilibrium in locations is... The market center: a=1-b=1/2 If there is no competition in prices firms tend to locate in the center of the variety space with a minimal product

differentiation.

Two Stages Game 1st Stage: firms choose their respective locations in a non-cooperative way. 2nd stage: firms choose their prices, given the location chosen in the first stage of the game, in a non-cooperative

way. We solve the game backwards: we start by analyzing the non-cooperative price equilibrium taking locationsas

given.Price equilibrium for given firms’ locations

P*A=t(1+(a-b)/3)P*B=t(1+(b-a)/3)

The pair (P*A, P*B) is a price equilibrium iff firms’ locations respecte the following conditions:

Price equilibrium for given firms’ locations If a=b (simmetric locations) then those conditions imply that a ≤ 1/4 e b ≤ 1/4

Quadratic Transportation Costs If we consider quadratic transportation costs we can show that each firm will choose to locate in one end of the

market.

Intuition: when firms are located far apart they reduce the intensity of price competition and gain market

power.

Topic 5 - Entry and Exit Entry BarriersStylized Fact:

Frequently we observe situations where there is no entry of new firms in the market despite the fact that incumbent firms make positive profits.

Question: why don’t we observe free entry until industry competition drives prices down and positive profits are eliminated?

Entry decision One firm decides to enter the market if the expected profits post-entry are higher than the sunk costs of entering

into that market (plants, machinery, patents, etc). The firm must anticipate the type of competition they will face after they enter into the market in order to

compute its post-entry expected profits. Entrants must take into account retaliation by incumbent firms.

Entry Barriers Bain: There are entry barriers if in the long run incumbent firms can set p>minAC without attracting new entries. Stigler: Entry barriers are defined as the costs that might be incurred by entrants but have already been incurred

by incumbents (cost asymmetry).Entry Barriers

Legal: markets with regulated entry (telecoms) Structural: incumbents have some advantages:

o costs (economies of scale and scope)o marketing (umbrella branding)o Control of critical resources (patent, mines) Ex: DeBeers (diamands), Alcoa (alumínium)

Strategic: results from incumbents’ entry deterrence strategies.o Capacity expansion,o Limit priceo Predatory pricing.

Markets classification according to entry barriers (I) Blocked entry: the incumbent firm does not need to adopt any strategy to deter entry. Ex: high structural barriers, economies of scale, homogeneous product (competition in prices is very aggressive)

Markets classification according to entry barriers (II) Accommodated entry: if structural barriers are low and

o Strategies for entry deterrence are not efficiento Or costs to deter entry are higher

In markets with fast growth or fast technological innovations entry is so attractive that incumbents cannot prevent firms to enter.

Caracterização dos mercados de acordo com as barreiras à entrada (III): Blockaded entry: the incumbent firm can use strategies to block entry. These strategies are successful if:

o The incumbent firm can increase price once it remains monopolist: It is not possible in perfectly contestable markets - hit-and-run strategy

o This strategy changes entrant expectations about the nature of post-entry competition.

Summary If entry costs are very high then the incumbent should set monopoly capacity and ignore the threat of entry. If entry costs are very low, then the incumbent should choose capacity, taking into account the entrant’s reaction

curve If entry costs are intermediate then the incumbent should choose capacity large enough to induce entrant not to

enter.Strategic Barriers (I)

Entry: we consider that entry occurs when a new firm starts to produce and to sell (is different from entry by acquisition)

Entrants threat incumbents in two ways:o They take way market share (reduction of the pie share)o Intensify competition (redction of the pie size):

Introduction of new products Prices decrease

Strategic Barriers (II) Limit Price: the incumbent firm deters entry setting a pre-entry low price

Objective: to make the entrant to believe that the post-entry price will be even lower and to persuade the firm against entry.

The incumbent can enjoy monopoly profits in the next future.Exemplo: P=100-Q, CF=850 por período, c=10, 2 períodos

Limit Price Failure: it is implicit that entrant has irrational expectations about the incumbent’s post-entry price policy. It works:

o If the incumbent has a cost advantage and sets P<minACEntrant

o If the entrant faces uncertainty about the rival’s cost or about demand.

Limit Price and Uncertainty:

Entrant’s uncertainty about incumbent’s costs Incumbent firm sets a price in the first period that corresponds to a low cost firm in order to discourage entrant to

entry the market. But this signal is not credible:

o A high cost firm may pretend that has low costs.o A low cost firm has interest in signaling his own type: sets a price that is not rational for a high cost firm.

Example with uncertainty

Example with uncertainty and Signaling

Predation If an incumbent monopolist cannot prevent entry it can still try to gain monopoly power by inducing the exit of its

rivals. Practices that have rivals’ exit as the main goal are known as predation.

Pricing below cost to injure rival firms and thus induce their exit is called predatory pricing. The incumbent sacrifices part of the profit that would be gain under normal competition conditions in order to

induce the exit of the rival and then bring price back to a high level and get the monopoly profits in the next periods.

Predatory Price: an irrational behavior? General Idea

o Predatory acts usually are not very successful because once the incumbent monopolist brings prices back to a high level, after a predatory pricing period; he attracts a new entrant again.

Predatory prices can be part of an equilibrium We must distinguish predatory prices from aggressive prices that are the result of a norma competition

framework.o Ex: Monopoly, homogeneous product, entry and Bertrand competition

The Long Purse (or deep pockets) A predatory pricing strategy can be successful if the prey is financially constrained and cannot stand a long price

war.Reputation

Frequently firms operate in different geographical markets Multi-product firms operate in different markets. Firms can act in a predatory way in a given market to signal potential entrants in other markets that they will fight

entry. Profit losses associated to a predatory strategy are the price payed to build a reputation of being “tough” so that in

the future (or in other markets) no entry will take place.Chain Store Paradox

Selten: if we have a repeated game where in each stage of the game there is a unique equilibrium of entry and accommodation, and if the game is played sequentially in different cities, is the incumbent’s interest to prey in some cities in order to build the reputation of being aggressive?

o Answer: noo Example: coffee wars in the US

Uncertainty If the entrant thinks there is a certain probability that the incumbent is “crazy”, i.e. she always prefers to fight

entry instead of accommodating, then we can have an equilibrium where the “non-crazy” incumbent can use a predatory strategy even in a single period game.

This strategy is used in the initial periods in order to create reputation of being aggressive.Fighting Brands

Quite often predatory pricing can be a too expensive strategy because it reduces prices for all units sold by the firm.

Fighting brands: the predator introduces a new brand specifically designed to compete with the prey’s product, keeping in this way his own brand “untouchable” and sets undercost prices.

Excess Capacity Credible threat of a price decrease if entry occurs Less suspicious for antitrust authorities Is an effective strategy even in a complete information setting. Investment in capacity must be sunk

Example: excess capacity to discourage entry (I)

Market where a monopolist faces entry threat Bertrand price competition and firms with fixed (K) capacity.

o Market demand: P=100-0.5(QI+QE)o TC= 100+30K+10Qo Q ≤ K

Example: excess capacity to discourage entry (II)

Example: excess capacity to discourage entry (III)

Exogenous sunk Costs ( J. Sutton’s model) Homogeneous product industriesSetup cost

Cost that a firm must incur when enters the market. It is the cost of setting up a unique plant with minimum efficient scale (minimum of average costs) net of resale

value. Since it is a sunk cost, the current price policy of the firm does not depend upon the magnitude of that sunk costs.

Firms’s decision: 1st Stage: firms decide about entry the market 2nd Stage: firms set prices Idea of long run and short run: Long run variables (entry) are decided in the first stage and are fixed parameters

in the second stage analysis when we compute the short run equilibrium (prices).Bain:

Rankings of industries by concentration level is quite similar across countries. Explanation: the technology pattern and consumers’ tastes (demand) that characterize an industry tend to be

similar across countries.Sutton:

Departs from the following observation:o In general industry concentration is lower in countries where the market size (sales volume) is high.

The Game

Prices depend on sunk costs level F but only in an indirect way because those costs influence the entry decision.Equilibrium Structure (Entry Decision)

1. Cournot Market Demand: Q=S/P P : price Q: Quantity S: Total expense Pº : monopoly price Hip.: Q=0 se P>Pº c: marginal cost, constant and identical for all firms.

Cournot Equilibrium

Entry decision (1st stage) Given rivals’ entry decision, a firm decides to enter to incur F and gets:

Elementary result When market size increases relatively to setup costs (S/F ) the market structure becomes more fragmented (N* ).

CONCENTRATION2. Bertrand

2 firms, identical costs, homogeneous product. Firms A is the incumbent; firm B is a potential entrant In case of entry firm B incurs a sunk cost F. If firms A is monopolist she gets PM- F If firm B enters the market the Bertrand equilibrium leads to P=cmg.

Price competition homogeneous product monopoly setup costs

3.Monopoly (Cartel) Joint profit maximization Joint profit is independent of the number of firms that entered the market Each firm’s profit depends upon N

Concentration and intensity of price competition

Conclusions:

There is a lower bound for the concentration level that can be sustained as an equilibrium The concentration decreases when market size increases

If two markets are similar in terms of setup costs (entry costs) but price competition is more intense in one of these markets then the lower bound for concentration in this industry is relatively higher.

Paradox: very intense price competition leads to a more concentrated equilibrium structure.Example:

Exogenous Factors that can change results Product differentiation Competition Policy

Example

Topic 6 – AdvertisingClassification of Goods

Search good: The consumer knows the good features before purchase (ex: computer). Experience good: the features of the good are known only after consumption (ex: red wine) This classification is important because the advertising effects can be different for each group of goods.

Advertising can be important for those consumers that want to buy experience goods because they need to gather information.

Classification of Advertising Expenditures Informative advertising:

Describes the product, its characteristics, pricing, selling points, etc... Persuasive Advertising:

Aims to change consumers’ preferences: our product is better, is safer... Makes the consumer to believe that a certain product is different. Deviates consumers from the rival firms’ products. (predatory) Expansion Effect vs. Competition effect

Stylized Facts The profitability of an industry is positively correlated to the advertising intensity in this industry. The advertising pattern of several industries remained relatively constant through time: industries that used to

have high advertising expenses in the 50’s, maintain these high levels of advertising expenses.Advertising as a “signal” for product quality

Is this a credible message? Firms with low quality products (LQP) can invest in adv. and sell at a high price to consumers that expect

to buy high quality products. Firms with high quality products (HQP) have more to gain from getting consumers to experiment their

products then LQP firms. HQP firms will receive repeated purchases

There is a level of advertising expenses that HQP firms are willing to incur but firms selling LQP are not: in equilibrium advertising is a signal of high quality.

Optimal level of advertising expenses Advertising has a positive effect on demand curve:

Dorfman-Steiner formula

The advertising-to-sales ratio is greater: the greater the advertising elasticity of demand the lower the price elasticity of demand (or the greater the price-cost margin)

Advertising: Optimal Values and Observed ValuesMarket h/e A/RInstant coffee 0,019 0,02Bottled beer 0,008 0,011Cigarettes 0,019 0,046Toilet soap 0,013 0,012Laundry washing powder 0,019 0,030Toothpaste 0,024 0,059

Fonte: Metwally, M. M. “Advertising and competitive behavior of Selected Australian Firms”, Review of Economics and Statistics, 47 (1975): 417-427

Dynamic Analysis The advertising impact on demand has an intertemporal effect:

The current advertising level affects de current demand but also affects the future demand (although with lower intensity).

Advertising increases the value of a long term asset – the brand – which affects the future demand.Advertising Dynamic Model

Let us consider a monopolist with a demand curve that depends upon price and goodwill stock – or brand. Qt=Q(Pt , Mt)

Where Mt is the brand value until the beginning of period t. The relationship between the advertising expenses and the brand value is:

Mt = Mt-1 (1-d)+ At Mt – Mt-1 = At -d Mt-1

Market Structure and Advertising Intensity Do firms advertise more in more concentrated industries? When the number of firms increases and concentration decreases there are 3 effects that determine the

advertising intensity: Each firm’s margin decreases (-) Each firm captures a lower share of the demand increasing effect of advertising – public good effect and

advertising spillovers (-) Each firm captures a larger share of the demand-shifting effect of advertising. (+) Net effect is ambiguous.

Empirical evidence: The advertising intensity increases until intermediate concentration levels and then starts to decrease.

Advertising and Price Competition Very often advertising leads to a decrease in price competition because it gives information about the product’s

characteristics and increases product differentiation. When advertising gives information about the products’ prices usually increases price competition.

Topic 7 - R&DR&D Interesting questions

Why are R&D expenditures so much higher in some industries than in others? Does industry structure have a significant effect on firms’ decision on R&D investment? What is the impact of R&D competition on market structure?

Ratio R&D Expenses/Sales

Innovation Discovery, development, adoption and trade of:

New process New product New organizational structure

Traditionally We assume that the production function is exogeneous to firms:

“black box” Introducing the R&D implies to consider changes in the firm’s production function.

Process innovation: investment to reduce the production costs Product innovation

Process Innovation Assumption: firms with the same technology compete à la Bertrand therefore MC=c0>0 Initially there is an unique equilibrium where all firms set price

P0 = c0 e

Assume that a firm invests in R&D and succeeds in decreasing production costs to c< c0. If PM(c ) is the monopoly price when unit cost is c then:

1. Innovation is non drastic if PM(c )>c0 (c0 >c1)2. Innovation is drastic if PM(c )<c0 (c0 >c2)

Non drastic innovation (1) Cost reduction is not large enough to permit the firm to set the monopoly price. Therefore the firm sets P1=c0-e <

c0, remains alone in the market and sells Q0. A small innovation

Does not change the market price Market sales remain the same.

Difference: in the present situation, the firm that has invested in R&D and has innovated keeps the entire market and makes a profit equal to (c0-c1)Q0

Process Innovation

Drastic Innovation (2) The innovation implies such a large cost reduction that the innovative firm is able to keep the entire market and

sets the monopoly price. A big innovation induces price reduction and increase in the quantity sold in the market.

What we can call a big or small innovation depends upon the market demand and the market structure, besides

the cost change in itself. Innovation Race

Timing of the innovation is crucial. The firm that innovates first a new process or a new product gains because:

Can obtain a patent and earn monopoly profits patent for several years; Consumers associate the innovator to a high quality producer and they are willing to pay for its product; Firms invest large amounts in order to be the first in the innovation race.

Questions Is the R&D investment level lower or higher than the social optimal investment level?

Spillover effects affect the optimal investment decision in R&D. Does market power reduce the incentives to innovate? Do firms in competitive industries have more incentives to engage in R&D?