Monopoly Practices 8 24 June 2019 - PowerPoint Presentation

1. Monopoly Practices824 June 20191

2. 8.1 Dominant-Firm Price Leadership Model25 June 201928.1.1 Sources of DominanceFew industries are truly monopolies. In practice, it is much more common to find industries in which there is a dominant firm as well some number of smaller fringe firms. Two factors can lead to a dominant firm–competitive fringe market structure:The dominant firm may have a significant cost advantage compared to its rivals, due to superior technology, management, or location or early entry and learning by doing, resulting in lower costs than rival firms. The dominant firm may have a superior product. For example, Intel’s microprocessors were traditionally the fastest and most advanced microprocessors available.

3. 8.1 Dominant-Firm Price Leadership Model25 June 201938.1.2 Pricing by a Dominant FirmDuring the first half of the twentieth century United States Steel (US Steel) was the dominant firm in the American steel industry, acting as the price leader for the smaller steel firms. In the figure, the steel industry demand curve D is (the dashed blue line for price between $25 and $100 and the black line for price below $25)The fringe’s supply curve is The dominant firm’s (US Steel’s) marginal cost curve is First find the residual demand curve for the dominant firm by subtracting the fringe supply curve from the total demand curve at every price greater than $25. for $25  P  $100 

4. 8.1 Dominant-Firm Price Leadership Model25 June 201948.1.2 Pricing by a Dominant FirmFor prices between 0 and $25, the fringe firms supply no output, so US Steel’s demand curve is identical to the industry demand curveUS Steel’s residual demand curve, therefore, has a kink at $25, and the MR curve has a gap at q = 75.To maximize profits, US Steel equates MR to MC, resulting in the profit-maximizing quantity for US Steel as 30 units.To obtain price, go up vertically in the to point Z on US Steel’s residual demand curve; P= $55.The fringe firms act as perfectly competitive price takers, supplying the quantity at which the fringe supply curve intersects the horizontal line P = $55, .Total industry output is the sum of US Steel’s output and the fringe’s output, 45.  

5. 8.1 Dominant-Firm Price Leadership Model25 June 201958.1.2 Pricing by a Dominant FirmOne of the major implications of the dominant-firm price leadership model is that the dominant firm’s market share declines continuously over time. If the competitive fringe firms earn above-normal economic profits, there will be an incentive for the fringe supply to increase over time as new firms enter and existing fringe firms expand output. As a result, the residual demand for the dominant firm will shift to the left (decrease), and the dominant firm’s relative share of output will decline.In the US Steel example, as new firms enter, the fringe supply curve rotates rightward, reducing residual demand for the dominant firm, which in turn reduces both market price and US Steel’s market share.

6. 8.1 Dominant-Firm Price Leadership Model25 June 201968.1.3 Empirical Evidence of the Decline of Dominant-Firm Price LeadersSome empirical evidence suggests that dominant firms’ market shares often decline substantially over time. US Steel behaved as a classic dominant firm after its formation in 1902. At the time of its creation, US Steel’s market share was 65 percent; by 1920 its share had declined to 50 percent. In the low-volume segment of the copier industry, Xerox behaved as a classic dominant-firm price leader, setting high profit-maximizing prices in this sector and conceded market share to its smaller rivals.Studies have examined whether deregulation resulted in markets that behaved according to the dominant firm price leadership model. Kahai, Kaserman, and Mayo whether AT&T was a dominant firm price leader in the telecommunications industry post-divestiture in 1984. Data covered the period 1984-1993 and considered if AT&T behaved as a dominant firm while Sprint and MCI were the primary members of the competitive fringe. From 1984 to 1993, AT&T’s market share in minutes-of-use market share declined from 84 percent to 59 percent as predicted by the model. Large increase in the number of competitive fringe firms over that decade.

7. 8.1 Dominant-Firm Price Leadership Model25 June 201978.1.3 Empirical Evidence of the Decline of Dominant-Firm Price LeadersA dramatic example is the case of the Reynolds International Pen Corporation, which invented an improved ball-point pen that operated on gravity. Reynolds began selling the pens in 1945. Initially the pens cost approximately 80 cents each to produce and Gimbel’s department stores sold them for $12.50. Gimbel’s sold 10,000 pens the first day they went on sale. By early 1946, Reynolds was producing 30,000 pens a day and earning large economic profits. But Reynolds’s high prices encouraged the competitive fringe to enter, and by Christmas 1946, many firms were in the industry and prices had fallen as low as 88 cents. By 1948 prices had declined further to 39 cents, and by 1951 they had declined still further, to as low as 19 cents. By then Reynolds was long gone from the industry. Points out clearly the risks associated with dominant-firm price leadership; by being completely passive toward its competitors, Reynolds found itself booted out of the market in just a few years.

8. 8.2 Contestable Markets: A Check on Market Power25 June 20198In 1982, Baumol, Panzar, and Willig advanced a new theoretical framework for thinking about entry and exit—the theory of contestable markets. The contestable markets hypothesis contends that potential competition may be more important than actual competition and that even a completely monopolized market may perform as if it were perfectly competitive in structure. Three basic assumptions:Entry is free. Free entry means that incumbent firms have no advantages, and there are no barriers to entry.Entry is absolute. If a potential entrant enters the market and charges a price below the incumbent’s price, then the entrant will completely displace the incumbent. No sunk costs are associated with entry. Permits “hit-and-run” entry in which firms can enter a market, extract profits for a period of time, and then withdraw with zero sunk cost losses. This assumption implies that a firm could, for example, sell its production facilities for their present value in a resale market or could use them in another market without any losses. It is important to note that this assumption does not eliminate fixed costs but only sunk costs.

9. 8.2 Contestable Markets: A Check on Market Power25 June 20199Consider the extreme case of a perfectly contestable market that is also a natural monopoly.Potential entrants exist. What price must the monopolist charge? If it charges the monopoly price and earns a positive economic profit, then the potential entrant will have an incentive to enter the market and charge a price below because by assumption it can enter with identical costs, displace the monopolist, and leave with no sunk cost losses.The only price that the monopolist can sustain is , the price at which the LRAC curve intersects the demand curve. A price equal to average cost is the only sustainable price because any higher price attracts entry, whereas any lower price results in economic losses.  

10. 8.2 Contestable Markets: A Check on Market Power25 June 201910Implies that to achieve a zero-profit result, all that is required is the existence of one potential entrant. In this model, potential competition is more important than actual competition, and even a natural monopolist may earn zero economic profits.Any movement away from the three basic assumptions would result in dramatically different results. If sunk costs were greater than zero, for example, then hit-and-run entry would be impossible and incumbents could earn excess profits without attracting entry.Baumol and colleagues pointed to the airline industry, which had recently been deregulated, as an example of a perfectly contestable market. This choice made some sense in 1982 because there had been a great wave of entry into the industry in the years 1979 to 1981. According to the proponents of the theory of contestability, entry into the airline industry required little more than a pilot and the leasing of an old plane. The firm could enter whatever market (flight between two cities) happened to yield the highest profit at the moment. In December, for example, the plane could fly from New York to Miami. If a World Series were being played between the Yankees and the Dodgers, the plane could simply fly back and forth between New York and Los Angeles. In this market, hit-and-run entry could truly evolve into a highly computerized art form. Any carrier charging a price above average cost would be displaced and new airlines would rule the skies.

11. 8.2 Contestable Markets: A Check on Market Power25 June 201911Is the airline industry a good example of a contestable market? Is entry free?While this might be true for equipment, potential entrants are likely to have difficulty obtaining gate slots for their planes due to long-term leases by airports to existing carriers. New carriers might have to spend significant amounts on advertising to penetrate any new market, while existing carriers already have brand name recognition. Is entry absolute?The assumption of complete displacement is highly questionable in any market, and is certainly not true in the airline industry, in which incumbents always have sufficient advance notice of any impending entry to permit them to respond with competitive price reductions. Are there zero sunk costs of entry?This assumption is refuted by the necessity of advertising to enter a new market, which establishes sunk costs for any entrant. None of the assumptions are valid for the airline industry, and thus airlines are not likely to behave in a way that is consistent with the theory of contestable markets.

12. 8.2 Contestable Markets: A Check on Market Power25 June 201912Anecdotally, the major airlines did not roll over and disappear when low-price carriers such as People Express, Midway, and Air West entered the industry. Better evidence is provided by a number of good statistical studies. Call and Keeler examined the effects of several variables on airline fares, using data on 89 city pairs. According to the contestable markets hypothesis, actual entry should have no effect on fares because potential entry should have already constrained fares to be equal to average costs. Similarly, the contestable markets hypothesis implies that concentration should have no significant effect on fares because, even in a monopolized market, fares should be forced down to average costs.Consequently, Call and Keeler’s test of the contestable markets hypothesis revolves around the statistical significance of three variables on fares: a measure of market concentration and two measures of entry. Their results show that all three variables have statistically significant effects on airline fares, strongly suggesting that concentration and entry have a significant impact on fares, which should not be the case if the market is contestable.

13. 8.2 Contestable Markets: A Check on Market Power25 June 201913Although the airline industry appears not to be a contestable market, the theory has had a major impact on the industry. Provided a rationale for the Reagan administration not to be concerned with concentration in the airline industry because, according to the theory, even a monopolized airline market would have to price at average cost. The contestable markets hypothesis with regard to the airline industry, therefore, was consistent with a policy that permitted virtually all mergers in the industry and resulted in higher market concentration.A few empirical studies have suggested that real world markets may exhibit contestability. Cowie: possible existence of contestable markets in the United Kingdom regional bus transportation market after the UK privatized the bus transportation industry.Corvoisier and Gropp, Molyneux, Thornton, and Lloyd-Williams suggest contestability in European and Japanese banking markets.However, in general, the theory of contestable markets has not been able to explain real-world phenomena well, although it may have uses in the field of international trade, where it might be used to explain the impact of potential foreign competition on domestic markets.

14. 8.3 Network Economics25 June 201914Recently, economists have realized that in markets for certain types of goods, network goods, a competitive equilibrium is unlikely to exist. Computer operating systemsSocial networking servicesAirline services, banking services, telecommunications servicesE-mail, the Internet, computer softwareOnline role-playing gamesMusic and video streaming services.Four main characteristics distinguish network industries from other industries:Complementarity, compatibility, and standardsConsumption externalitiesSwitching costs and lock-inSignificant economies of scale in production

15. 8.3 Network Economics26 June 2019158.3.2 Complementarity, Compatibility, and StandardsSome products must be consumed together with other products (complements): Digital cameras are not useful without memory cards, a DVD player is used in combination with DVDs, and a computer and software are used together. In order to work together, components must exhibit compatibility, which implies that they must operate on the same standard. In 1860, there were more than 30,000 miles of railroad track in the United States, but there were at least five different gauges (distance between rails) of track, which made it difficult and expensive to transfer traffic from one railroad to another.A more modern example would be accessories for an iPod, which need to be compatibleWhen compatibility is required, some coordination is necessary to make sure that components work together, and this may be difficult. There are transactions costs associated with communicating across firmsThere may be concerns about violating antitrust laws by sharing information. Thus a firm may choose to produce multiple components itself. Combined with consumers’ tendency to shop for systems, this can increase the entry barriers for new firms.

16. 8.3 Network Economics26 June 2019168.3.2 ExternalitiesThe value of goods or services produced in network industries to a potential customer depends on the number of consumers already owning that good or using that service. This kind of side effect is an externality, and the externalities arising from network effects are referred to as adoption or network externalities.Can be positive: Facebook is more valuable because many people use itPositive network externalities can result from bandwagon effects: a consumer may place a higher value on a good because others have it, such as trendy clothing.Can be negative: more users can create congestion or interferenceNegative externalities can also result from snob effects: a consumer may value exclusive or unique goods more than those that everyone has.In the case of a direct network externality, the benefit to an individual consumer depends directly on the number of other users of the product.The utility of a social network depends on the number of your friends also on the networkA phone provides no utility if no one else has a phone.

17. 8.3 Network Economics26 June 2019178.3.2 ExternalitiesIndirect network externalities arise from complementarity.The value of a game system such as PlayStation depends in part on the set of games available to play on the system. As more consumers purchase a particular system, more games will be developed for that system. Network externalities affect the adoption of a new technology. Early on, only a few consumers will subscribe to a new service or purchase a new good. However, after a specific percentage of consumers, called the critical mass, has subscribed to the service or bought the product, its value increases to become greater than or equal to the price. After this point, all potential consumers subscribe to the service or buy the good.Example: E-mail Although the first e-mail was sent in 1969, it wasn’t until sometime in the mid-1980s that it began to be widely used. Now nearly everyone uses e-mail

18. 8.3 Network Economics26 June 2019188.3.2 ExternalitiesIn the presence of network externalities, an early entrant often gains a competitive advantage. For example, consider Mirabilis, the Israeli start-up that developed instant messaging (IM). Mirabilis gave its product away for free to increase the number of users. Mirabilis also prevented interoperability between its client software and other products to deny competitors access to its customers. By using these strategies, Mirabilis was able to dominate the market for IM. Network externalities played an important role: New IM users were much more likely to join Mirabilis’s system than a competing system.

19. 8.3 Network Economics26 June 2019198.3.3 Switching Costs and Lock-InMost users don’t like learning features of a new operating system, a new word-processing program, or a new spreadsheet. Similar switching costs arise in service industries such as banking. Switching costs can arise in a variety of forms, including the costs of breaking a contract, the costs of training and learning a new system, the costs of converting data from one software program to another, the costs of searching, and the costs associated with giving up loyalty benefits such as frequent-flyer miles. In the presence of significant switching costs, consumers are locked in to using a specific service or product. Once this occurs, firms know that they have some room to raise price without losing consumers. Since firms understand the value of locking in consumers, they may use strategies such as discounts and free services to attract consumers who are not yet locked in. Explains the fact that competition for consumers can be intense even in a market where firms have market power.

20. 8.3 Network Economics26 June 2019208.3.4 Significant Economies of ScaleProduction of goods or services in network industries often involves significant economies of scale. A new computer game takes hundreds or thousands of hours of programming time to develop, and the costs of this development time are sunk. Once the program has been written, the marginal cost of producing an additional copy (on a CD or on a website to be downloaded) is almost zero. This type of cost structure—very high fixed sunk cost combined with very low marginal cost—exists whenever the product involves information. Economies of scale may also be associated with the central node in a network. Adding more consumers to an electricity network allows the fixed costs of the central generating facility to be spread over more users. The presence of significant economies of scale often leads to a market dominated by one or a few firms.

21. 8.3 Network Economics26 June 2019218.3.5 Math of Network ExternalitiesA relatively simple model, drawn from an early analysis of network economics by Rohlfs, shows some key points. In order to focus on the demand-side issues, the model assumes that the market is a monopoly. Think of a telecommunications firm, and further assume that the firm charges a single price, p, for being connected to the network but that each individual call is free.There are N potential customers in the market; since the number of customers is fixed, can also consider the fraction, f, of the market that that is served at any price.An individual consumer’s demand for connecting to the communications service depends on two factors: his or her valuation of the product and the fraction f of the market that has subscribed. Assume that the network externality is positive.But individual valuations, denoted by for the ith consumer, also differ.To separate these two factors, consider a case in which everyone subscribes (f =1). Assume that the valuations with f = 1 are uniformly distributed between $0 and $200.  

22. 8.3 Network Economics26 June 2019228.3.5 Math of Network ExternalitiesThe next step is to show the effect of the network externality on a consumer’s utility. A simple way is to model the utility of the ith customer as A consumer’s utility depends not only on his or her individual valuation but also on the fraction of the market actually subscribing to the service. Consumer i will subscribe to the network if his or her utility is greater than or equal to the market price: . Demand by the ith consumer will be 0 if . Consider the marginal consumer—the consumer who is just indifferent between joining the network and not joining it. Denote the valuation of this consumer by Because this consumer is indifferent, we know that the valuation is just equal to the market price: , or We also know that any consumer with a lower valuation will not subscribeThe assumption of uniform distribution implies that the fraction of consumers in this group of non-subscribers is  

23. 8.3 Network Economics26 June 2019238.3.5 Math of Network ExternalitiesFor example, suppose the marginal consumer has a valuation of $50. In this case, ¼ of the consumers in the market will value the service at less than $50 and will not subscribe.The number who will subscribe is , and so ¾ of the consumers will join the network. Substituting for the valuation of the marginal consumer from yields which can be solved for p:Shows the relationship between the market price and the fraction of consumers who subscribe to the service. Graphing this gives the hump-shaped relationship shown in the figure.  

24. 8.3 Network Economics26 June 2019248.3.5 Math of Network ExternalitiesWith positive network externalities, the demand curve is upward sloping.The network externality pulls the reservation price upward as more consumers join the system. Can lead to two equilibrium points, one unstable and one stable. For example, a price of $37.50 results in equilibrium points 1 and 2 in the figure, with only the equilibrium at point 2 with 75 percent of the consumers joining the network being stable. This shape of a demand curve is characteristic for consumer choice in the presence of network externalities. Beyond f*, the size of network associated with the peak reservation price, the demand relationship is as expected for any product: there is an inverse relationship between price and the number of consumers so that a provider has to decrease price in order to attract additional consumers. Below f*, there is a positive relationship: the reservation price increases as more consumers join the network. In this region, the positive network externality is pulling the reservation price up even though the valuation that marginal consumer places upon the product decreases as network size increases .

25. 8.3 Network Economics26 June 2019258.3.5 Math of Network ExternalitiesThe figure also shows the multiplicity of equilibria in this type of market. At any price , there is a “low-demand” and a “high-demand” equilibrium. At point 1, only consumers who value the service highly will join the network. Using the example, if price is $37.50, the low-demand equilibrium is f = 0.25. The low-demand equilibrium is unstable: once a network reaches this size, positive feedback from users leads to expansion of the network. One more consumer joining the network at this price would increase the value of the service above $37.50 for all consumers in the interval from 1 to point 2. Eventually, as these users subscribe, the network reaches point 2, the high-demand equilibrium, a stable equilibrium. In our example, f = 0.75 at point 2. Rohlfs called the low-demand equilibrium the critical mass of consumers; it represents the minimum number of subscribers the provider must enroll in order to have a non-zero demand.  

26. 8.3 Network Economics26 June 2019268.3.5 Math of Network ExternalitiesAnother common characteristic of consumer demand under network effects is coordination problemsIn order to decide which type spreadsheet software (e.g., Excel, AppleWorks, or IBM Lotus Symphony, a consumer needs to know how many other users there are. Think about the possibility of having no subscribers at a given price, which is a stable equilibrium, since no one would want to be the only person who uses a type of spreadsheet. The fact that a zero-demand equilibrium is stable means that a provider of a good with network externalities has to think about strategies to start the network and get the number of subscribers to increase to the critical mass. Such strategies might include: Giving the product away, which happened in the early years of competition in the market for dial-up Internet Service Providers (ISPs) lead by America-on-Line (AOL) and Internet browsers led by Microsoft’s Internet ExplorerPaying off your contract with another mobile telecommunications company like T- MobileDramatically reducing prices as happened in the market for spreadsheet software where prices declined by over 50 percent from 1986 to 1991.

27. 8.3 Network Economics26 June 2019278.3.6 Summary of Network EffectsCompetition among products is often unstable when network externalities are strong. Larger networks have an advantage which increases as new users join and as consumers switch from competing products. This bandwagon/tipping effect makes it difficult for many smaller networks to stay in business unless they have important distinguishing characteristics or have been able to lock in their users.Because of the bandwagon/tipping effect, competition among rival products in the early stages is likely to be cutthroat. Producers understand how important it is to attract customers, making strategies like low introductory pricing and claims about future product developments appealing. Early behavior can lead to an anti-competitive market structure, but information in early stages is incomplete and often confusing. Policymakers want to avoid interventions that work against technological improvements.

28. 8.3 Network Economics26 June 2019288.3.6 Summary of Network EffectsEntry in network industries is likely to be more difficult than entry in non-network industries that are similar in other ways. It is hard to convince consumers to try a new product or service when they would have to incur switching costs as well as give up the benefits of the network externalities. Entry can occur if a new product or network is sufficiently attractive, but the costs and risks are higher. Raises extra concerns for antitrust policy.Network effects do not have to lead to one firm dominating the market. A key is whether standards are established that allow multiple firms to produce competing products. For example, strong competition exists within the market for long-distance telephone service within the United States. No one firm has a proprietary protocol that it can use to dominate the market. However, in other cases, network effects can be used to enhance monopoly power (See Application on Microsoft).