Graduate School of Business



Graduate School of Business

University of Chicago

Business 33001 Robert Topel

Autumn 2009

Lecture 9

Strategic Behavior I: Monopoly

I. Introduction: What This Lecture is About

This lecture studies strategic behavior by sellers. To this point we have generally assumed that sellers are price takers—there are so many sellers in a market that no single one of them can affect price. This assumption is often unrealistic. Who believes that Microsoft cannot affect the price of its software products, that Toyota cannot affect the price of its cars, or that Coca-Cola cannot affect the price of cola? For that matter, who believes that the local movie theater cannot choose its price of admission or that the local pharmacy would sell no toothbrushes if it slightly increased their price?

Sometimes the assumption of price taking behavior is a useful simplification that makes our analysis much easier. To analyze the impact of a tax on movie tickets, we can safely ignore the fact that each theater has some control over its price. We assume that each theater is a price taker, and the supply/demand framework of earlier Lectures will do nicely. But in other cases the assumption that all sellers are price takers is a hindrance that would keep us from understanding important aspects of a problem. If we want to understand why theaters give discounts to senior citizens, it does us no good to assume that there is only one price in the market, and that each seller takes it as given.

This lecture deals with situations in which sellers set price in order to maximize profits, so we call them price setters. Price setting behavior has many important implications. One is that markets may not be efficient. Price is generally too high, and the quantity sold is too low. This fact helps us to rationalize some government interventions that might make markets more efficient, such as price regulation of public utilities and antitrust policies that combat monopolization. Another implication is that firms may have complex pricing strategies. For example, they may charge different prices to different buyers, as when airline charge different fares for seats on the same flight, or when colleges charge higher tuition to the children of wealthy parents.

Lecture 9 deals with the implications of price setting behavior when there is only one seller, called a monopoly, in the market—though even in this case the definition of the “market” can be a little dicey. We begin with the case where the monopoly sells all its output for the same price, called single price monopoly, and we show how such a seller chooses its price and quantity sold. We then establish that single price monopoly is inefficient—price is too high and quantity too low compared to the socially efficient outcome, and we discuss the costs (and benefits) of monopoly.

II. Monopoly

Monopoly is Greek for “one seller.”

Monopoly is the polar opposite of perfect competition, and you might think it is a rare thing. Think about the products you buy. For how many of them is there really only one seller? Not many. But there are a few obvious cases of monopoly for goods and services that we purchase every day. For example, there is only one seller of electricity in the City of Chicago, Commonwealth Edison. We say that Edison has a monopoly in the sale of electricity to residential customers in the City of Chicago. Other “public utilities” that sell gas and water are also monopolies because they are the only sellers from whom we may buy those products.

These are obvious cases, but there are others of great practical importance. In the recent antitrust litigation against Microsoft, the U.S. Court of Appeals found that Microsoft is a monopoly in the sale of operating systems for Intel-compatible personal computers. Yes, there are other operating system products (Linux and some others) that can be used on the same Intel hardware, and consumers could get almost exactly the same functionality if they bought an Apple. But the court found that, for practical purposes, Microsoft was a monopolist in this “market.”

Price Setting: The Practical Importance of Monopoly

The most important practical implication of monopoly is that a monopolist has some control over the price at which it sells. The left-hand graph in Figure 9.1 shows the market demand for electricity in metropolitan Chicago, DE. Since Commonwealth Edison is the only seller in this market, the demand curve for Edison’s electricity is exactly the same as the market demand. Then if Edison raises its price from [pic]to [pic], consumers will buy [pic] fewer units of electricity from Edison.

The conclusion that consumers buy less electricity if Edison raises its price seems reasonable, but notice the difference between Edison and a price taker. The demand curve for Farmer Jones’ wheat is shown in the middle graph of Figure 9.1. Jones is a price taker, with no control over price, so the demand for his wheat is perfectly elastic at the market price, [pic]. He can charge [pic] for his wheat and sell as much as he wants, but if he charges more no one will buy from him. Unlike the demand for Jones’ wheat, the demand for Edison’s electricity slopes down—it isn’t perfectly elastic—because it is the same as the market demand.

The practical importance of this is that Edison does not take the price of electricity as given. Instead, Edison must choose a price at which to sell its electricity. (Whether the government will allow Edison to charge the price that Edison prefers is another matter. The issue of regulating Edison’s price is taken up below.)

Definition: Price Setting Behavior

A seller that can choose its price is called a price setter. A price setter faces a downward sloping demand curve for its product, while a price taker faces a horizontal (perfectly elastic) demand curve for its product.

A price setting firm is often said to have market power because it has some control, or power, over the price at which its sells.

[pic]

Product Differentiation and Market Power

Jones can’t raise the price of his wheat because his wheat is just like anyone else’s, and he accounts for a trivial share of total wheat production. If he raises price, people will buy the identical product from someone else. The demand for Jones’ wheat is perfectly elastic, which demonstrates a general proposition: a seller’s ability to raise price is smaller the greater is the elasticity of demand for its product. As we learned in Lectures 2 and 3, availability of substitutes is one of the key determinants of the elasticity of demand for a product. The elasticity of demand for Jones’ wheat is perfectly elastic because many other farmers produce wheat that is a perfect substitute for his. In contrast, consumers in Chicago have no good substitute for Edison’s electricity. If Edison raises its price consumers cannot easily switch to other sellers.

In the real world, true price takers like Jones are the exception rather than the rule. A little refection should convince you that most sellers have some control over the price at which they sell, so they are price setters with market power. Harry the local barber will get less business if he raises the price of his haircuts, but not all his customers will leave him. Carlo, a modern artist, would sell fewer paintings if he raised his price, but he will still have buyers. The owner of a small Chinese restaurant thinks long and hard about the price she should charge for a meal, and whether she should raise it on the weekend. And when Apple introduced its IPod music player in 2002, management also thought long and hard about the best price to charge.

Example: In the original DOJ antitrust suit against Microsoft, Microsoft’s internal documents recounted a debate among executives about the price to charge for the newest version of Windows. Microsoft calculated that it would be profitable to sell Windows for about $45, but it would be more profitable to sell it for $80. The judge took this internal debate as evidence of Microsoft’s monopoly power.

Does this mean that barbers, artists, restaurant owners, and Apple Computer are all monopolists? In a sense, yes. We said that monopoly occurs when there is a single seller in the market. And, strictly speaking, in a market all sellers must offer an identical product. In the examples we mentioned, each of the sellers offers a product that is a little different than anyone else’s. In other words, there are no perfect substitutes for these goods, so they are not literally price takers.

For example, Harry’s is the only barbershop in his neighborhood. Other barbers produce haircuts that are good substitutes for Harry’s, but so long as local customers value the convenience of a short walk to the barbershop, a haircut sold by another shop two miles away is not a perfect substitute for one sold by Harry. This product differentiation means that haircuts produced by Harry are a unique good, and so we can (loosely) think of a distinct “market” for them. Harry faces a downward-sloping though highly elastic demand curve, as shown in Figure 9.1c. He has some “market power,” but not much.

Definition: Product Differentiation

Product differentiation means that a seller’s product has unique characteristics that are valued by buyers. Sellers of differentiated products face downward sloping demand, so they are price setters.

Product differentiation can take many forms. Harry’s customers value the convenience of his location. The Porsche Boxster has a unique design. Viagra is a chemical compound, patented by Pfizer and valued by users, that may not be copied by other pharmaceutical manufacturers for the life of the patent. In each case, product differentiation gives a seller some control over price. Our next task is to show how price setters choose the most profitable combination of price and quantity.

Single Price Monopoly

Jose makes great teriyaki armadillo at his Tex-Mex restaurant. He uses a secret recipe that he got from his grandmother. No other teriyaki armadillo is quite like Jose’s, they say, which gives Jose some control over the price he charges for an armadillo dinner—he is a price setter. The daily demand for teriyaki armadillo dinners is linear, as shown in Figure 9.2:

[pic]

If Jose wishes to sell just one armadillo dinner per day, he can charge a price of P(1)=$21. If he wants to sell 2, however, we assume that he must cut his price by $1 on both of the dinners he sells. If he wants to sell 3, he must cut his price to $20 on all three dinners, not just the third. We say that Jose is a single price monopoly because he must sell every armadillo dinner for the same price. Later we will take up the situation where a monopolist can charge different prices for different units, called multi-price monopoly.

[pic]

Table 9.2 records information on Jose’s demand, revenues, costs and profits. Each armadillo dinner costs Jose $6 to produce—ingredients are cheap because armadillos are road-kill in Jose’s area. So his costs are [pic] and marginal cost is [pic] per dinner. The table shows that Jose’s profits are maximized at $56 per day when he produces 8 dinners per day.[1] At this rate of output he charges $13 per armadillo, so his “profit margin” (price minus marginal cost) is $13-$6=$7 per armadillo.

|Table 9.1 |

|Daily Demand, Revenue, and Profit Data for Jose’s |

|Tex-Mex Teriyaki Armadillo Dinners |

|[pic] |

|(1) |

|Quantity |

|[pic] |

| |Word |Excel |Office |

|Poets |$300 |$150 |$450 |

|Geeks |$200 |$250 |$450 |

|Optimal Price |PW = $200 |PE = $150 |PO = $450 |

|Revenue |$400 |$300 |$900 |

|Consumer Surplus |$100 |$100 |$0 |

Now suppose Word and Excel are bundled together into a new product called Office, and are not sold separately. Poets are willing to pay $300 + $150 = $450 for Office. So are Geeks. So a price of PO = $450 for Office will just attract both of them. Total revenue is 2x$450 = $900, and neither the Poet nor the Geek gets any consumer surplus. Microsoft extracted all consumer surplus by bundling the two goods. So in this example the creation of Office enabled Microsoft to engage in perfect price discrimination.

Interpersonal Price Discrimination

Another form of price discrimination, which we see almost every day, is charging different prices to different consumers. This is called interpersonal price discrimination. Look back at Homer’s demand curve in Figure 9.8, and pretend that each bar represents the willingness to pay for one beer by separate customers who, like Homer, have their marginal values tattooed on their foreheads. Then Crusty could earn a profit of $25 by charging customer #1 $12 for his beer, customer #2 $10, and so on out to customer #5. To do this, Crusty had to (1) be able to identify which person was willing to pay more, and (2) prevent resale of beers from the person who got a low price (#5) to those who are willing to pay more. These are basic conditions that must be satisfied:

Conditions for Interpersonal Price Discrimination:

In order to charge different prices to different customers for reasons unrelated to cost, a seller must:

1) Separate buyers into classes according to willingness to pay.

2) Prevent resale from those who are charged a low price to those who would be charged a high price.

A particularly stark example of these conditions arose many years ago for a manufacturer of plaster. The plaster was particularly suited to making dental impressions, so dentists were willing to pay a lot for the stuff. But it could also be used for plastering walls, so builders and home re-modelers also bought the plaster, though they were willing to pay less because there were substitutes available. So the maker of the plaster charged a higher price to dentists than for bulk purchases from industrial supply stores. But the dentists figured out that they could go to the building supply store and buy plaster for much less, which undermined the price discrimination scheme—the seller could not “prevent resale.” To eliminate the problem, the plaster manufacturer mixed arsenic in the plaster sold to building supply stores, with a skull and cross-bones on the bag. That prevented resale.

This was an example of a firm actually incurring a cost to charge a lower price. It is more expensive to make the plaster with arsenic than without it, and the arsenic-laced plaster was sold for less. There are many other examples of this. The cost to make an Intel Celeron chip is actually greater than for Intel’s higher-priced chip—it’s just the high-priced chip with some functionalities shut down. The Celeron sells for less, to people who don’t demand as much computing power from their PCs. Hewlett-Packard did the same thing with calculators. HP sold machines to scientific and high-end business users for a high price, and less technically advanced machines to students and others for a lower price. But the low-priced machine was basically the same as the high-priced one, with some functions shut down. It actually costs more to “turn off” the functions, but without doing so HP could not prevent scientists from buying the cheap version. So it was profitable for HP to incur a cost in order to sell something for less. Mercedes and BMW sell cars that are wired for built-in cell phones, but the functionality is turned off unless you are willing to pay $1500 to enable it.

Examples of interpersonal price discrimination are ubiquitous. Here are some:

Airline Tickets: Major airlines employ armies of people whose sole job is to figure out how much we are willing to pay, and make us pay it. So if I will stay over a Saturday night in Wichita, I get a lower price. Why? The person who will spend Saturday night in Wichita is more likely to be visiting grandma. Those trips are more discretionary, and people are willing to pay less for the trip. The traveler who really wants to be home on Friday night is likely a business traveler, who is willing to pay more. Again, the airlines don’t know exactly who will pay the most, but they divide customers into classes and the customers identify themselves by their behavior.

College Financial Aid: The U of C and other select universities have a list price (tuition) that is currently over $30,000 per year. The average amount actually paid is about half that, however, as most students get some form of financial aid (i.e. a discount). How do we decide who gets a lower price? To get a scholarship, parents must submit their income tax forms for several years, along with an accounting of their assets. “How much do you make? Send it in. Thank you.” (Yes, we actually punish people who save, because they have higher ability to pay.) This example is a good one because it illustrates the problem of (1) dividing customers into groups according to willingness to pay—the rich will pay more, and they do; and (2) preventing resale. It isn’t possible to sell your degree, even if you got it cheap, to someone else. Education is a service, which can’t be resold—I can’t sell my education or my haircut to someone else. This is why price discrimination is common in services. Doctors charge higher prices to rich patients. Golf instructors charge more to rich clients. And so on.

Sales: Large department stores have regular “sale” periods when prices of merchandise are temporarily reduced. For example, Marshall Fields has an annual “13-hour” sale when most goods in the store are discounted. (Like the example of Celeron chips, it costs money to re-price stuff). Why? People who want a new pair of shoes now are willing to pay more for them than people who are willing to wait for the shoes to go on sale. By having the sale, the low-value customers wait for the low price, while the high-value customers buy for the higher price.

One way of characterizing the persons who purchase for a lower price is that they are more elastic demanders of the good, so a successful interpersonal price discrimination scheme charges a higher price to the more inelastic demanders of the good. To see this, suppose that a seller can divide buyers into two groups, the Hs and the Ls. Assume that the marginal cost of selling to each group is c per unit. Then a profit maximizing monopolist that can prevent resale will maximize profit by setting marginal revenue equal to marginal cost in each “market”, H and L:

[pic]

Then

[pic]

In other words, the price discriminating seller charges a higher price to the more inelastic demanders of the good—if the L’s are more elastic demanders, they get a lower price. Here is a good example.

Wholesale Pricing of Brand-Name Drugs: Pharmaceutical companies sell their products to pharmacies, who then sell the drugs to consumers who get prescriptions from doctors. Historically, consumers obtained their drugs from local pharmacies where a pharmacist filled the prescription while the customer waited. This practice changed with the emergence of mail order pharmacies such as Medco and Caremark in the 1980s. Consumers with chronic ailments, who required the same drug month-after-month, could obtain their drugs by mail from large, centralized distributors. Customers mailed in the prescription, which was kept on file, and Medco mailed the drug.

This technology created new opportunities for substitution among “therapeutically equivalent” drugs. Zantac and Tagamet are therapeutically equivalent drugs used to treat ulcers because they have the same effect for most people. Viagra, Levitra and Cialis are therapeutic equivalents. There are many other examples. When a patient sends a Zantac prescription to Medco, Medco pharmacists have the time to call the prescribing doctor and urge a switch to, say, Tagamet. This opportunity to substitute was not present for the local pharmacist, who did not have the scale or the time to make such a call while the customer was waiting. As we know, one of the determinants of the elasticity of demand is the ability to substitute, so the Medco’s practice make the elasticity of demand for Zantac and Tagamet greater in mail order than at local pharmacies. Medco used this as a bargaining tool. If Smith-Kline would give Medco discounts on Tagamet, then Medco would make Tagamet its preferred ulcer drug. When a Zantac prescription came in, they would attempt to convince the doctor to switch, arguing that a switch would save the patient money. More often than not the doctor agreed, so Medco’s strategy was successful. Smith-Kline and other drug companies granted discounts to mail order pharmacies and others with the ability to substitute.

In this example, and all of the others mentioned above, the lower price goes to the more elastic demander of the good. So when you are deciding whether a price difference reflects price discrimination, you need to convince yourself of three things. First, does the price difference simply reflect a difference in cost of selling to one group over another? French wine sells for more in the US than in France, but that would be true even with perfect competition in the market for French wine—the price difference reflects the cost of transporting the wine to the US. The price difference does not reflect price discrimination.

Second, is the seller able to prevent resale of the good? If it seems that resale is cheap and easy, then the price difference probably reflects differences in costs, as just mentioned, rather than price discrimination.

Third, is it plausible that the customers paying a higher price are the more elastic demanders of the good? If not, the difference in prices is unlikely to reflect price discrimination.

III. Summary and Conclusions

“Monopoly” is a term used to describe a market with only one seller. In this sense monopoly is the opposite of perfect competition, where there are so many sellers than no one of them can materially affect the market price. Yet in practice the line between competition and monopoly is more blurred. Most firms, even in industries that are highly competitive, have some control over price. This means that for some problems like our analysis of price discrimination the competitive model is not a useful tool. The monopoly model provides useful insights that help us understand real world behavior, even when it is applied in seemingly competitive market situations.

The most important implication of this Lecture is that monopoly is inefficient. Compared to the efficient outcome, a monopolist will restrict output in order to raise price. The ability to raise price, called monopoly power, is limited by the elasticity of demand for the seller’s output. When demand is highly elastic, price will be close to marginal cost and the inefficiency of monopoly pricing will be small.

[pic]

-----------------------

[1] He could earn the same profit by producing 7 dinners per day. So his marginal profit on the 8th dinner is zero. We assume for now that Jose will produce and sell a dinner so long as his gain in revenue is at least equal to the cost of producing it. The fact that two quantities yield the same maximal profit is due to the discrete nature of the data—if armadillos were infinitely divisible, like applesauce, then the profit-maximizing output would be unique.

[2] Joseph Schumpeter, Capitalism, Socialism, and Democracy.

-----------------------

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

The demand for Edison’s electricity =market demand

The demand for Jones’ wheat

The demand for Harry’s haircuts

Figure 9.1

Price Setters and Price Takers

[pic]

[pic]

[pic]

Figure 9.2

Daily Demand for Jose’s Teriyaki Armadillo Dinners

[pic]

[pic]

Figure 9.4

Profit Maximizing Price and Output for a Single Price Monopoly

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

Figure 9.3

Demand, Marginal Revenue and Elasticity

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

Figure 9.5

Profit Maximizing Price and Output for a Single Price Monopoly: Diminishing Returns

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

Figure 9.7

Profit Maximizing Price and Output for a Single Price Monopoly: Increasing Returns to Scale

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

Figure 9.6

Lost Consumer Surplus: Monopoly Profit and

Deadweight Loss

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

Figure 9.8

Homer’s Demand for Beer at Crusty’s Tavern

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

[pic]

Figure 9.9

A Two-Part Pricing Scheme

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download