The American Antitrust Institute today endorsed a government proposal to get tough on predatory practices at air transportation hubs. In comments filed with the Department of Transportation, the AAI commended a D.O.T. proposal to institute guidelines for recognizing and prosecuting unfair exclusionary behavior by major carriers1. D.O.T. is responding to allegations that major carriers maintain their dominance of hubs by cutting prices and taking other actions intended to knock out new entrants. After the entrant departs the market, the major carrier resumes its previous pricing. The AAI commented that it is appropriate under the principles of antitrust and in the interest of consumers to promote more competition at air hubs by vigorously attacking predation.
D.O.T.'s proposals have been heavily criticized by the established airlines. The AAI responds, "The adamant opposition of the established large carriers in the press and on Capitol Hill is a reflection of the potential efficacy of the Department's plan to reduce predatory attacks on new competition. Their suggestions that an effort to curtail predation will deter competitors from aggressive price competition, to the detriment of consumers, conveniently ignore the longer term impact of predation, namely the elimination of aggressive price competition."
AAI recognized that the D.O.T. tests are in certain respects vague, and it therefore recommends the inclusion of a "safe harbor" for major carriers, which would take them out from under the proposed rules provided that they commit to maintaining any strategic price reductions on a long-term basis. AAI President Albert A. Foer explains, " Under our suggestion, dominant carriers could have certainty in return for agreeing to hold their price reductions in place for a period of time, such as two years, even if the new entrant departs the market. If they choose not to use the safe harbor option, they will be implicitly admitting that their price reductions are intended to be temporary, only for the purpose of ultimately excluding competition."
The American Antitrust Institute is an independent non-profit organization, recently created in Washington, D.C., to promote the objectives of the nation's antitrust laws.
The American Antitrust Institute's
Comments to the Department of Transportation
Enforcement Policy Regarding Unfair Exclusionary Conduct in the Air Transportation Industry
Docket No. OST-98-3713
Statement of Interest
The American Antitrust Institute is an independent non-profit organization whose purpose is to promote the objectives of the nation's antitrust laws.
The deregulation of air transportation in 1978 was premised on the assumption that a competitive market would adequately protect consumers, without the necessity for burdensome economic regulation by the C.A.B. The implicit understanding was that vigorous antitrust enforcement would assure a competitive market that would serve consumers better than regulation. Unfortunately, antitrust oversight of the air transportation industry has not fulfilled its end of the bargain, with the result that we now have an industry that is concentrated in too few players, whose dominance of various system hubs results in self-regulation by monopolists rather than regulation by competitive markets. With the movement toward domestic airline alliances further reducing the level of competition in our air transportation industry, it is all the more important to focus on increasing competition at the hubs.
We are pleased that D.O.T. is addressing the problem through this proposed enforcement policy. In these comments, we support the thrust of the proposal and suggest, in addition, an optional safe harbor whereby dominant carriers can avoid any uncertainty implicit in the D.O.T. tests by committing to maintaining in place for a defined period of time any price reduction made in response to new entry.
The Nature of the Problem
The D.O.T. proposal responds to widespread allegations of predation. Obviously, there will have to be evidence to sustain a reasonable belief in the allegations, but the allegations essentially involve the emergence of a pattern in which major carriers become dominant at so-called hub air terminals, and that they maintain their dominance by strategies which have the intent and result of excluding new entrants.
These strategies involve both pricing and non-pricing elements. The pricing elements include reducing prices for seating on routes that are being challenged by new entrants; and offering frequent flier bonuses to consumers who fly with them on the challenged routes. The frequent flier benefits have value to the consumer and therefore can be viewed as a form of additional discount. Non-price strategies include adding additional seats at low fares; scheduling flights to bracket the times offered by the entrant; and aggressive marketing targeted against the entrant. The elements of strategy are often combined, and the overall objective is to keep consumers from switching to the entrant.
Once this strategy succeeds and the entrant withdraws from the market, the elements of the strategy may be withdrawn, and, in particular, the "fighting fares" will be phased out. Empirical investigation can determine whether it represents a common pattern and whether additional entrants indeed continually reappear, as the Chicago School theory discussed below would suggest, or whether a dominant carrier, once it has demonstrated its intent and ability to crush a newcomer, can successfully phase out its combat strategy and comfortably resume highly profitable operation at the hub. Our comments assume that the latter situation will be confirmed.
Predation and Antitrust
The D.O.T. proposal focuses on the identification of predatory practices by dominant air carriers, aimed at keeping new entrants from encroaching on their captive hubs. Before commenting on the proposed approach, it may be useful to place the D.O.T. proposal into the historical context of antitrust's treatment of predatory unilateral business practices.
In the early days of America's experiment with antitrust, predatory acts by a dominant firm were of great concern to the public and to the law. The landmark case that resulted in the breakup of the Standard Oil trust in 1911 was premised on evidence that Standard Oil had engaged in a variety of predatory acts intended to kill off its smaller competitors2. Predation remained a major concern of antitrust, but scholars identified with the "Chicago School" of economics began in the 1970's to question the logic of predation3. Working from static analyses based in price theory, they reasoned that it might not make sense for a firm to engage in predatory behavior. The logic goes like this:
Firm A has substantial market power in market X, and observes Firm B entering the market. Firm A responds by reducing its prices to the point where it cannot make a profit, or even loses money on each unit it sells, with the intent that it will drive out Firm B, recapture its dominance, and raise its prices. It will only do this, says the "Chicago School", if it foresees that it can sustain its eventual elevated prices at least long enough to recoup the investment made in predation (i.e., recover the profits lost by selling cheap). Unless there are barriers to entry, once Firm A raises its prices, however, Firm C will make the decision to enter market X, and Firm A will have to make yet another investment in predation. Since the "Chicago School" does not believe that entry is generally difficult, in the absence of government-created barriers, their conclusion is that in general, corporate strategies of price predation will not be worth pursuing. Moreover, as a matter of public policy, they argue, we would not want to deter vigorous price competition, which is in the interests of consumers. We should opt not to risk intervening in most situations where competitors are alleging that they are being damaged by predation: since predation is unlikely to occur, our interventions would have an unnecessarily chilling effect upon true price competition.
This line of argument was enhanced by a vigorous academic dispute over the definition of price predation. How should a court or an enforcement agency draw the line between prices that are merely aggressive, and those that are likely to be predatory? The most famous proposal in this area was developed by professors Areeda and Turner, who said that only a price that is the below the firm's own costs should be deemed predatory, and that cost should be defined operationally in terms of average variable cost. 4
The high water mark of the "Chicago" line is the Supreme Court's opinion in Matsutshita Electric Industrial Co. v. Zenith Radio Corp.5 in 1986, a case in which dumping charges were brought by American TV manufacturers against Japanese manufacturers. The Court cited the Chicago School literature on predation and suggested that predatory pricing would only harm consumers in particular market environments where recoupment of losses through raised supracompetitive prices was likely. The Court made it clear that it thought price predation rarely occurs and is extremely unlikely. This was expanded in Brooke Group Ltd. V. Brown & Williamson Tobacco Corp.6 , where the Court held that in a predatory pricing case, a plaintiff must prove (a) that the prices complained of are below an appropriate measure of costs, and (b) that the alleged predator had a reasonable prospect, or a "dangerous probability," of recouping its investment in below-cost prices.
In the face of the influential "Chicago School" critique of predation, a "Post-Chicago" movement has emerged, which has answered the critique and developed deeper insights into how competition works both in the real world and as a matter of theory.7
First, the Post-Chicago rejoinder says, it is fundamentally wrong to think that a static analysis provides an adequate picture of how firms compete. When Firm A commits itself to the investment in keeping Firm B out of Market X, it is sending a message not only to Firm B but to other firms that might contemplate entry into Market X. Moreover, if Firm A also operates in other markets, it may be sending a message to other companies it faces in other markets. The message is: "I am one tough, aggressive warrior; if you cross me, I will do whatever is in my power to crush you." In other words, the investment is in building a particular kind of reputation that is thought to have strategic value in the overall business of the firm. It does not necessarily have to be recouped in Market X.
But, second, even if the predatory investment does have to be recouped in Market X, the analysis has to turn to the thought processes of the next firm that might contemplate its own possible entrance into Market X. It has to ask itself (or its investors will ask), in light of what happened to Firm B when it tried to enter, will Firm A maintain its demonstrated aggressive policy8? Chicago School economists believe that entry is generally easy, but they tend to overlook the sunk costs that are a part of most market entry strategies. In reality, the next potential entrant is likely to be deterred if it believes that Firm A has a predatory character and the market power to predate.
Third, the Post-Chicago rejoinder has argued that a focus solely on price is in many situations too difficult to apply. Obtaining clear-cut information about a firm's variable costs with regard to one of its many products not only requires access to detailed accounting data, but also requires second-guessing and debating a host of accounting decisions relating to the proper allocation of expenses. In practice, cost-based tests of predation have been difficult to apply, even when there is agreement on the appropriate test.9 A better test might focus on the firm's strategic intent, which is admittedly less quantifiable than price/cost data, but is not necessarily any less valid.10
And fourth, the Post-Chicago rejoinder has noted that price is only one aspect of a strategy. Various forms of non-price predation have been identified and analyzed, some focused on strategies aimed at imposing losses on rivals, others at raising rivals' costs. For example, it has been theorized that firms may sometimes change a product's characteristics in order to head off entry by a rival11 or that entrenched firms may try to make it more expensive for firms to enter or introduce new products by increasing the costs of acquiring market information, a tactic which has been dubbed signal jamming.12
What this brief history of predation and antitrust means for the D.O.T. effort to deal with allegations of predatory behavior at air transportation hubs may be summarized as follows:
- While the D.O.T. statutory authority for regulating unfair exclusionary practices is not identical to the Sherman Act, the objectives are similar and the nation's experience with predation under the antitrust laws should provide a useful context for the D.O.T. However, D.O.T. has the flexibility and expertise to determine what is predatory in the air transportation industry's particular circumstances and to develop appropriate guidelines to avoid predation.
- The D.O.T. should be attempting to provide guidelines that will promote aggressive competition that places emphasis on assuring that new entrants have the opportunity to enter hubs, so that they will be able to assure consumers the fruits of competition, including long-term low fares.
- This implies that both the pricing and non-pricing strategies of dominant carriers must be understood from three perspectives: first, what is intended by the dominant carriers? Second, how are the major carrier's strategies likely to interact with the entry strategies of rivalrous carriers? And third, how will any remedy affect consumers?
- The D.O.T. need not focus entirely on a cost-based definition of predation, which may prove too difficult to administer and may not be sufficiently encompassing to deal with the strategies that are at work in the air hubs. The antitrust community has learned a lot from the Areeda-Turner debates over cost-based definitions of predation, and has been sensitized to the need to avoid rules which will inhibit aggressive competition; but it has also learned that competition, as studied in the business schools and played out in the marketplace, is a rich and diverse art form, not easily captured by a single school of economic theory.
Comments on the D.O.T. Proposal
The D.O.T. Proposal contains three tests of unfair exclusionary conduct in response to new entry, any one of which can result in the institution of enforcement proceedings. The tests may be summarized as (1) adding capacity and selling a large number of seats at very low fares; (2) carrying more passengers at the entrant's low fares than the entrant's total seat capacity; or (3) carrying more passengers at the entrant's low fares than the number of low-fare passengers carried by the new entrant. In each test, the major carrier would have to lose more local revenue by self-diversion than would be lost by a reasonable alternative strategy. Also, enforcement proceedings will not be instituted if there are strong reasons to believe that a major carrier's response to competition from a new entrant does not violate 49 U.S.C. 41712.
To be effective, an approach of this sort must work quickly and the penalties must be decisive. An entrant who is driven from the market long before the administrative process has brought relief may well be a dead entrant. Penalties that amount to a slap on the wrist will not deter major carriers from following a strategy that has the desired outcome. Therefore, the proposal should include provisions which make it more likely that a new entrant will be protected in a timely fashion, if protection is called for, and that the major carrier will have a strong incentive to avoid strategies that are unfairly exclusionary.
The approach taken in the D.O.T.'s Proposal leaves many things vague, as indeed may be necessary when we talk about a variety of possible strategies and seek a balance between the goals of boundless aggressive competition and rigorous-but-fair competition. For example, much rides on what will be determined to be a "reasonable alternative response", a phrase in the D.O.T. test whose meaning will vary with the circumstances. The more that rides on the elucidation of this phrase (i.e., the greater the penalties), the more prolonged will be the administrative process leading to a determination. It is especially important to an entrant to be have greater certainty in predicting the costs of entry. Rather than try to provide a more simplistic test, however, we suggest the addition of a safe harbor option which may appeal to some major carriers while providing greater predictability to all: if the major carrier chooses to anchor within the safe harbor, the three D.O.T. tests will not apply while the carrier is in the harbor.
A Safe Harbor Proposal: Commitment to Maintaining Price Reductions
Our suggestion is this: a safe harbor will be recognized if a major carrier (defined in terms of having at least 50%13 of the seat capacity or the passengers flown within a specific hub air terminal) acts in response to the public announcement of a new entry14(which is simultaneous with or followed by actual entry), by committing to maintaining its price reductions for a specified number of seats for a period of two years15 , without regard to whether the entrant remains in the market.
Commitment would take the form of a publicly filed notice to the D.O.T. The major carrier would be required to file public interim reports demonstrating that it had maintained its commitment. The D.O.T. would be expected to investigate claims that the commitment was not being honored and violation of the commitment would constitute a per se unfair exclusionary practice. A major carrier would be entitled to seek a semi-automatic increase in its commitment fare on the basis of passing through increased costs of doing local business. It would also be entitled to request other increases on the basis of good cause demonstrated. (However, exit by the entrant carrier would not constitute good cause.)
This proposal is grounded in analysis originally presented by the economist William J. Baumol.16 Baumol noted the difficulty in applying cost-based definitions of predatory pricing, and attempted to set a rule that would be "relatively easy to administer, permits full and fair competition by both entrants and established firms, and encourages enduring, rather than temporary, price cutting, in order to serve consumers' interests over the long run."17 It leaves the established firm free to cut prices in order to protect its interests, without being permitted to re-raise those prices if the entrant leaves the market or if the firm wants to subsidize price cuts of other products that are then threatened by competition.
In other words, a quasi-permanent pricing requirement has the benefit of not raising a protective umbrella over the entrant. While the established firm is free to respond, it will no longer be free to respond without fear of long-term repercussions. It will adjust its price if the value of the resulting competitive gain will exceed the long-run cost of a long-term reduction in price.18
From the perspective of a major carrier, the strategic question of how to respond to a new entrant would change from
How long will it take to force B out of Market X if we match or undercut his low fares?
What price reduction are we willing to live with for two years, regardless of whether B leaves the market?
We will address five questions that present themselves. First, what happens if the major carrier makes a reduction, and then sees that it is insufficient? In the interests of flexibility and low prices for consumers, additional reductions may be made, but each constitutes a new commitment, starting the time period again.
Second, what happens if the price selected is truly below cost (however defined)? Should this be permitted within the safe harbor? On the affirmative side, the entrant will know whether he can operate at a profit in the context of the major carrier's committed price. If not, he knows what he will be up against and he can exit quickly, leaving the major carrier to suffer losses for the two- year period. Administratively, there is benefit in not having to investigate the relation of price to cost. If the major carrier's prices are below the new entrant's costs, the public will spend less and consumers will benefit, depending on how many seats are available at the low fare. On the negative side, the entrant might have provided more seats at low fares than the major carrier does under its commitment; and, of course, the entrant will find it very difficult to remain in the market unless its own costs are lower than the major's deep discount price. On balance, we think it most likely that a major carrier, pondering the costs of selling a competitively significant number of seats below cost for two years, is more likely to follow a less costly strategy and so we would permit flexibility without setting a lower limit.
Third, how would multiple pricing points and seasonal variation in rates be recognized? Since the proposal is couched in terms of a number of seats (not per flight or per time slot, but per route in total) for which a reduced price will be semi-permanent, it would be acceptable if a major carrier established price discounts for various buckets. Further, it would be acceptable if the prices were to apply to defined seasons. For example, a major carrier could respond to a new entrant by announcing that it was reducing prices for X number of seats in three different pricing categories on a particular route and that these reductions would hold true for the months of April through July, while other rates would hold true for August through November, etc.
Fourth, what happens at the end of the two year period? If the major carrier is no longer dominant (over 50%), then of course the rules will not apply. Moreover, it should be assumed that two years is long enough for the particular entrant to gain a foothold and compete on a level playing field.19 Thus, neither the D.O.T. tests nor the safe harbor provision would apply to the major carrier that had graduated from the safe harbor. However, if another carrier seeks entry, the tests would apply, and the safe harbor provision would be available.
And, fifth, it is one thing to assume that the new entrant is a small carrier or even a start-up airline. What if it is a large, well-established carrier that is attempting to gain entry? The safe harbor approach would be equally applicable. In determining how it will respond to a new entrant, the major carrier can take the characteristics of the newcomer into account. Because a large carrier would have the deep pockets to withstand a price war, it is less likely that the major carrier would start a price war. Which just makes the point, that a temporary price war is more likely to be waged against a small carrier that is unlikely to be able to withstand a predatory strategy. It follows that if a major carrier creates a "fighting brand" division or subsidiary or joint venture operating under a different name, such carrier will not be deemed a new entrant; rather, it will stand in the shoes of the parent and be considered a major carrier, regardless of its market share.
The American Antitrust Institute commends the D.O.T. approach as an appropropriate but incomplete effort to deal with predatory strategies that appear to be keeping our hub terminals safe from the penetration of new competition. The adamant opposition of the established large carriers in the press and on Capitol Hill is a reflection of the potential efficacy of the plan in reducing predatory attacks on new competition. Their suggestions that an effort to curtail predation will deter competitors from aggressive price competition, to the detriment of consumers, conveniently ignore the longer term impact of predation, namely the elimination of aggressive price competition. We recommend that the D.O.T. approach be supplemented by the stated intention to apply strategically significant penalties for unfair exclusionary practices and that a safe harbor based on quasi-permanent price reductions be included.
Albert A. Foer
The American Antitrust Institute
1 Docket OST-98-3713, "Enforcement Policy Regarding Unfair Exclusionary Conduct in the Air Transportation Industry".
2 Standard Oil Co. of New Jersey v. United States, 221 US 1 (1911).
3 See, e.g., Robert H. Bork, The Antitrust Paradox (1978), and Yale Brozen (ed.), The Competitive Economy (1975), which contains a leading article by John S. McGee, "Predatory Price Cutting: The Standard Oil (N..J.) Case," that questions the logic and occurrence of price predation.
4 Phillip Areeda and Donald F. Turner, "Predatory Pricing and Related Practices under Section 2 of the Sherman Act," 88 Harvard Law Review 697-733 (1975). Average variable costs are the measure most commonly used by courts, although there are variations among the federal circuit courts.
5 475 U.S. 574 594 (1986).
6 509 U.S. 209 (1993). For a narrow reading of the implications of this case, see Jonathan B. Baker, "Predatory Pricing After Brooke Group: An Economic Perspective," 62 Antitrust Law Journal 585 (1994).
7 A recent layman's summary of these developments may be found in the May 2, 1998 issue of The Economist at pp. 62-64, concluding, that the Post-Chicago theories "will motivate enforcers to investigate business behavior that hitherto would have raised no eyebrows. They will come to understand new ways in which businesses acquire excessive market power. Consumers should be grateful."
8 For a while in the 1980's, there was a Chicagoan theory called "contestable markets" which argued that even a monopolist has to set prices as if it were up against competition, because raising prices higher than the competitive level would induce "hit and run" entry by potential competitors who would come in, make their profit, and then exit if the monopolist responded with reduced prices. "Contestable markets" theory was shot down because it ignored the role of sunk costs and exaggerated the ease of entering a market. The theory now carries little weight in that it would be so rare to find a market where it could apply.
9 See, e.g., Oliver E. Williamson, Antitrust Economics, 225 (1987).
10 Richard Posner , in Antitrust Law, An Economic Perspective (1976), suggested that both intent and cost need to be considered. His definition of predatory pricing is: pricing at a level calculated to exclude from the market an equally or more efficient competitor. Id. at 188.
11 See Thomas J. Campbell, "Predation and Competition in Antitrust: The Case of Nonfungible Goods," 87 Columbia Law Review 1625 (Dec. 1987) and John C. Hilke and Phillip B. Nelson, "Nonprice Predation and Attempted Monopolization: The Coffee (General Foods) Case (1984) in John E. Kwoka, Jr., and Lawrence J. White, The Antitrust Revolution 208 ( 1989).
12 See Steven C. Salop and David T. Scheffman, "Raising Rivals' Costs," 73 American Economic Review 267 (May, 1983); Thomas G. Krattenmaker and Steven C. Salop, "Anticompetitive Exclusion: Raising Rivals' Costs to Achieve Power over Price," 96 Yale Law Journal 209 (Nov., 1986); Steven C. Salop and David T. Scheffman, "Cost-Raising Strategies," 36 Journal of Industrial Economics 19 (Sept., 1987); Drew Fudenberg and Jean Tirole, "A 'Signal Jamming' Theory of Predation," 17 Rand Journal of Economics 366 (1986).
13 This represents a low-side estimate of the market share required for dominance. D.O.T. should select a threshold that complies with its understanding of industry dynamics.
14 In order to avoid gaming by potential entrants, we suggest that if new entry is not followed by actual entry, the commitment would be cancelled.
15 Two years is used here for purposes of elucidation. A reasonable way to establish a time period would be to investigate situations at hubs where the major carriers reacted by cutting prices and the entrant was forced to withdraw. On average, how long did it take to convince the entrant to withdraw? Using the antitrust analogy of treble damages, triple that time. The objective is to make it more expensive to adopt a strategy of predation by assuring that the steepest price reductions (which are most likely to succeed in forcing out an entrant in the short term) are the most expensive for a predator.
16 William J. Baumol, "Quasi-Permanence of Price Reductions: A Policy for Prevention of Predatory Pricing," 89 Yale Law Journal 1 (Nov., 1979). Baumol suggested that five years might be a reasonable term for a quasi-permanent price reduction, but noted the need for "the light of experience and considerations of practicality."Id. at 8.
17 Id., at 2.
18 While a major carrier would be free to ignore the safe harbor option, it could find itself in the awkward position of having to explain why it is non-predatory to cut prices against a new competitor only so long as the entrant remains in the market.
19 The D.O.T. proposal does not deal with that happens down the road after the new entrant is no longer new.