ABA Section of Antitrust Law and Section of International Law

December 2006

(ABAComments.pdf ((PDF: 77 KB)))

The American Bar Association Section of Antitrust Law and Section of International Law (together, the "Sections") appreciate the opportunity to submit these Joint Comments on the Canadian Competition Bureau's (the "Bureau") September 26, 2006 Draft Information Bulletin on the Abuse of Dominance Provisions as Applied to the Telecommunications Industry (the "Draft Bulletin"). These comments present the views of both Sections as approved by their respective Councils. ¹ Neither the House of Delegates nor the Board of Governors of the American Bar Association has approved these comments and they do not represent the policy of the American Bar Association.

Discussion

The Sections support the Bureau's preparation and circulation of the Draft Bulletin for comment as a means of increasing efficiency, consistency, and transparency in the Bureau's process of reviewing cases for potential abuse of dominance by firms operating in the telecommunications industry. The Sections understand, however, that the institutional and regulatory framework governing the telecommunications industry in Canada remains a topic of continued discussion and that the current leading case under the abuse of dominance provisions is subject to a leave to appeal application to the Supreme Court of Canada. Accordingly, the Sections recognize that future developments may require future revisions to the Draft Bulletin.

Our comments address specific paragraphs of the Draft Bulletin under five topics: (1) market definition, (2) market power assessment, (3) anticompetitive acts, (4) the substantial lessening or prevention of competition, and (5) remedies.

I. Market Definition

The Bureau should indicate that the Draft Bulletin applies across the telecommunications industry to both business and residential services, as well as to both retail and wholesale services. The Draft Bulletin currently focuses on residential telecommunications services.² Abuse of dominance claims, however, may also arise in business-related telecommunications services.

II. Market Power Assessment

A. Role of Market Power in Abuse of Dominance Cases

The test for dominance set forth in the Draft Bulletin involves an assessment of whether a firm or group of firms has the ability to act independently of competitive discipline, to a material degree, with respect to price, output, quality, variety, service, or innovation. The Draft Bulletin would benefit from more specific references to relevant Canadian case law.

B. Market Share

The Draft Bulletin identifies market shares as an important indirect indicator of market power. The Sections recognize the Draft Bulletin provides that high market shares are viewed as a generally necessary, but not sufficient, condition for the existence of dominance and that without barriers to entry any attempt by a firm with a high market share to exercise market power would be met by entry and expansion of competitors. Section 3.2 could state more specifically that the Bureau will not treat market shares as conclusive evidence of dominance.³

In addition, we note that footnote 36 explains that the Bureau will continue an examination of a matter if it finds that the market share of a party exceeds 35%. This level of concentration may be too low for a presumption of market power, particularly in a recently deregulated market where market shares may merely be a vestige of regulation and not a gauge of market power.⁴ Setting this threshold too low may cause firms with modest market shares to expend a disproportionate amount of management time and financial resources to monitor their compliance with Canada's dominance laws. Accordingly, the Bureau may want to articulate other factors it will consider in addition to market share in deciding whether to continue an examination.

The Bureau also may wish to provide more practical advice to businesses by considering using an appropriate market share-based safe harbor below which dominance would not be found.⁵

C. Barriers to Entry

The Draft Bulletin appropriately notes that in the absence of entry barriers, a firm with high market shares will not be able to increase prices or constrain industry output. This section lists economies of scale, sunk costs, regulatory barriers, long term contracts, market maturity, reputation of incumbents, and network effects as entry barriers. The Sections have the following comments on this list.

1. Intellectual Property

Considering the importance of intellectual property to the telecommunications industry, the Bureau should consider adding a discussion on the role of intellectual property in creating or maintaining entry barriers. The Bureau's Enforcement Guidelines for the Abuse of Dominance Provisions contains a discussion of the role process patents played in the NutraSweet case.⁶ A similar discussion of the importance of intellectual property in telecommunications in this section seems appropriate.

2. Network Effects

The discussion of network effects could benefit from additional explanation. The Draft Bulletin explains that network effects occur where the value of a product or service increases as more customers use that product or service, e.g., as more customers join a network. Whether network effects act as barriers to entry depends on the circumstances.

If competing networks are compatible and able to interconnect, network effects generally are less likely to act as an entry barrier. Where users of one network can seamlessly interface with users of another, the sum of users on those two networks is what determines demand-side value to consumers. For example network effects may not be a major factor for VoIP service providers if participants on these networks derive value from their ability to participate in communications with participants on the wireline local telephone network.

Network effects may be more likely to act as an entry barrier for compatible networks where the number of participants on the network affects the quality of service. New compatible networks may be accessible only within a limited geographic area, while service within that area may suffer from a lack of service infrastructure due to limited participation on the network. Service on the incumbent network also could suffer from network effects where too much network participation reduces the quality of service offered to other network participants, for example, through the high occurrence of dropped calls on a wireless network. Whether network effects act as a barrier or actually contribute to ease of entry in compatible network markets depends on the factual circumstances in that particular market.

Network effects are also relevant where networks are incompatible, for example, in new networks that depend on disruptive technologies that do not interface with existing networks. Considering the value proposition facing telecommunications firms, one would expect most new networks to strive for compatibility with existing networks and for existing networks to strive to maintain compatibility with competing networks.⁷

3. Period of Price Impact

The Draft Bulletin suggests that market power cannot be exercised where timely entry is likely. In footnote 39 the Draft Bulletin notes that normally the impact of entry on prices must occur during a two-year period to be timely. However, the Bureau's Enforcement Guidelines on the Abuse of Dominance Provisions uses a one-year period for assessing whether a firm has market power. The Draft Bulletin should explain whether a one or two year period will be used for assessing an exercise of market power in the telecommunications industry.

4. Supply-side Substitution

The Draft Bulletin addresses supply-side substitution in the context of entry barriers. However, the Bureau's Merger Enforcement Guidelines include firms capable of supply-side substitution without significant sunk costs or distribution supply issues within the relevant market, presumably for the purpose of calculating market shares. We suggest that the analytical approaches in the Merger Enforcement Guidelines and the Draft Bulletin should be consistent, even if the analytical differences are not likely to produce substantive differences.

5. Mature Markets

We question the presumption that only mature markets are difficult to enter.⁸ It may be that where legacy infrastructures or other advantages are used to leverage existing market power into new or emerging markets entry into an "emerging market" may be similarly difficult. The Draft Bulletin should recognize the risk that a dominant operator's control over infrastructure bottlenecks supported by its former monopoly position could prevent competitors from developing new and emerging services at the retail level.

D. Countervailing Buyer Power

We have several comments on the Draft Bulletin's discussion of countervailing buyer power, i.e., a customer's ability to constrain the exercise of market power. In many cases, countervailing buyer power can be an effective constraint to the exercise of market power, but some cases require a more comprehensive analysis. The mere presence of certain strong buyers who may be able to extract more favorable conditions from the allegedly dominant undertakings may not prevent a price discriminating supplier from exercising market power with some identifiable class of customers, because: (1) the class of identifiable customers who may be discriminated against does not have countervailing strategies available to them; (2) the cost of the product or service may be a relatively insignificant part of the overall cost of the end product, and the strong buyer, for strategic purposes, including reciprocal dealing, may not have the incentive to exercise its countervailing market power; or (3) dominant undertakings have strategies to minimize buyer power, including the use of conditional rebates, that could be used to lock in sufficient demand so that remaining providers could not realize minimum economies of scale and scope. In these circumstances even strong buyers may not be able to exercise any countervailing power by threatening to take their business elsewhere or to self-provision.

III. Anticompetitive Acts

A. Raising Rivals' Costs and Market Foreclosures

The Sections caution the Bureau to avoid treating raising rivals' costs as synonymous with reducing rivals' revenues. Raising rivals' costs does imply reducing rivals' profits, while reducing rivals' revenues may be consistent with lower prices and hence with competition. The Draft Bulletin appropriately distinguishes between the ability of dominant telecommunications firms to raise costs of facilities-based rivals and their ability to raise costs of rivals that require access to the dominant firm's facilities. While a dominant telecommunications firm may be able to raise facilities-based rivals' costs, as in the Draft Bulletin's example of foreclosing access to efficient inputs or means of distribution, facilities-based firms have different abilities to respond to dominant firm conduct with different effects on competition.

B. Margin Squeezing

The Draft Bulletin states that the Bureau "will consider whether [a margin squeeze] is based on justifiable business practices that would be found in otherwise competitive markets."⁹ The general and unspecific terms "justifiable business practices" and "otherwise competitive markets" do not afford firms guidance in developing their business strategies. Under this language a dominant firm would be subject to the Bureau's questioning regarding virtually any decision to lower its retail prices, potentially chilling highly desirable conduct.

In the Draft Bulletin, the Bureau provides an example where "a vertically integrated dominant firm sells its wholesale service to both competitors and large business end-users."¹⁰ The Draft Bulletin states that as a result of this action, "an increase in the wholesale price may compete with it in the downstream (retail) market."¹¹ Specifically, it argues that "this would be the case if an examination of the evidence shows that the primary reason for the price increase related to increased costs associated with serving large business end-users, even though that sector does not compete in retail markets."¹²

This example uses an objective, economic standard, namely: whether the dominant firm's retail price is below an appropriate measure of the dominant firm's costs, which costs would incorporate the wholesale price charged rivals for the input that is the subject of the margin-squeezing claim. Antitrust enforcers should hesitate to challenge price-reducing strategies by a dominant firm unless that firm is pricing below marginal cost because pricing at marginal cost is the very definition of competition.¹³

C. Denial of Access to a Facility

The Draft Bulletin defines "essential facility" as "an input that provides the firm controlling it with the power to lessen or prevent competition in a relevant downstream market."¹⁴ Defining "essential" in terms of harm to competition does not provide guidance that would enable a dominant firm to identify the kinds of facilities for which the denial of access will raise concerns with the Bureau. Inquiring whether denial of access to competitors would give the dominant firm power over price or would enhance its power over price would provide a more specific standard.

The Draft Bulletin proposes three elements for an essential facilities claim: (1) "[a] vertically integrated firm that is dominant in two markets," (2) "denial of access to the facility for the purpose of excluding competitors from entering or expanding in the downstream market or otherwise negatively affecting their ability to compete," and (3) a substantial lessening of competition.¹⁵

1. Dual Market Dominance

The Draft Bulletin does not explain why the dominant firm in the market for the alleged essential facility must be dominant in the downstream market as well. This requirement may be both over inclusive (false positives) and under inclusive (false negatives). A firm has no anticompetitive motive to extend dominance into the second market if the upstream and downstream products/services are used in fixed proportions, because that situation implies the availability of only one monopoly rent.¹⁶ In addition, where both stages of production exhibit imperfect competition, the upstream dominant firm's integration into the downstream market may eliminate supra-competitive markups charged by downstream firms, which would be an efficient outcome.¹⁷

The Draft Bulletin's test also may lead to false negatives. A dominant firm may realize greater revenues from downstream customers switching to it from the downstream competitors than it could realize from providing access to the facility for downstream competitors, if that firm could expand downstream capacity sufficiently to accommodate the additional customers. The dominant firm also may be able to dominate the downstream market quickly because of low switching costs for downstream customers, low marginal costs of output expansion, or network effects, all of which are characteristic of telecommunications markets and not indicators of anticompetitive conduct. The Bureau should evaluate each situation according to its own economic circumstances.¹⁸

2. Purpose or Intent

The Draft Bulletin's second element, the "purpose of excluding competitors from entering or expanding in the downstream market or otherwise negatively affecting their ability to compete,"¹⁹ is difficult to apply. "The defendant's intention is seldom illuminating, because every firm that denies its facilities to rivals does so to limit competition with itself and increase its profits."²⁰ The Bureau also notes that, "for the purpose to be anticompetitive, the supplier must have the necessary capacity, or have the willingness and ability to build the necessary capacity, to supply those competitors."²¹ A firm's inability to grant access to competitors is a legitimate justification and thus, an appropriate prerequisite to finding anticompetitive intent.

However, the concern should not be entirely on the effect on particular competitors, but rather the harm to competition generally, and the effect this has on consumers. Thus, output expanding conduct that may provide cost advantages that are passed on to consumers is positive, even if it means a particular competitor is less able to compete.

3. Lessening Competition

The Sections commend the Bureau for grounding its analysis in whether the denial of access substantially lessens competition. This test more closely approximates the U.S. Supreme Court's decision in Trinko, which held that a unilateral refusal to deal can give rise to antitrust liability only if there is a dangerous probability of monopolization in the downstream market.²²

D. Predatory Pricing

The Sections agree with the Draft Bulletin's position of avoiding overzealous enforcement against predatory pricing and applying "a high evidentiary requirement" to pursue predation cases. In particular, the Draft Bulletin recognizes: (1) the difficulties associated with distinguishing instances of true predatory conduct from competitive behavior and (2) the risk that wrongfully pursuing procompetitive behavior will constrain beneficial pricing. However, the statement, "[t]he Bureau is not likely to pursue a complaint where the alleged predator earns profits in that market at the alleged predatory price" suggests that the Bureau may consider pursuing above-cost predatory pricing claims. The Sections caution against such claims and suggest that the Bureau provide a clearer statement in the Draft Bulletin that above-cost pricing is not predatory.

The Sections agree in general with the Draft Bulletin's three step approach, which first determines whether a predatory price would be likely to drive a firm from the market, then determines whether the price is below average avoidable cost, and finally determines whether the firm would recoup the profit sacrifice predation would entail. However, the Bureau might consider augmenting its approach.

1. Cost Standard

The Draft Bulletin adopts the average avoidable cost standard as its benchmark for suspiciously low pricing, rather than the average variable cost standard articulated by the late Professor Areeda and adopted by many U.S. courts.²³ The Sections recognize that the Bureau has used the avoidable cost standard in past cases but believe the average variable cost standard is worthy of careful consideration,²⁴ may be easier to implement in practice, and is more consistent with the financial data used by most firms to run their businesses.

These standards raise two related questions: (1) Over what period must price be below the appropriate measure of cost to trigger suspicion of predation? (2) What volume of sales is required to trigger intervention? The Bureau might consider that the period in which a firm assesses the profitability of its commercial strategies may extend beyond the period in which the supposed predation takes place. For instance, an internet service provider may incur initial losses with new clients that may be compensated over the average "client life." Such situations require close examination of the particular details presented.

2. Cross Subsidy

The Sections recommend that the Bureau identify the market characteristics under which a cross subsidy may provide a firm subject to rate regulation with "an unusually credible threat to predate in [the] unregulated markets [in which it participates]." The ability to cross subsidize does not necessarily increase the threat of predation Cross subsidization occurs when a firm is able to maintain or increase profits by shifting common costs from an unregulated product to a product that is price regulated. Such shifting increases the rate base of the regulated product, resulting in higher prices since the regulated prices are generally set on a cost-plus basis. Cross subsidization allows firms to make greater profits in the regulated industry irrespective of pricing in the unregulated market. Thus, a profit maximizing firm will operate at a level where costs are allocated to maximize profits across the regulated and unregulated industries, regardless of whether it seeks to prey. Accordingly, in equilibrium, a profit maximizing firm will be unable to lower prices in the unregulated industry and maintain profits through additional cross subsidization, unless something peculiar about the rate regulation increases cost shifting to the rate base during episodes of low pricing in the unregulated sector. An explanation of how the presence of rate regulation and cross subsidies increases the chances of recoupment would provide more useful guidance to firms using cross subsidies.

3. Business Justifications

Given the potential harm to competition of misinterpreting certain competitive conduct as predatory, the Sections suggest that the Bureau always evaluate the existence of alternative business rationales such as minimizing losses, meeting competition, efficiencies, and promotional price-cutting. One objective justification for allowing prices below average avoidable cost arises when the dominant firm is minimizing losses in the short run. In addition, the introduction of new products or the introduction of new features in a given product may justify such a pricing policy where costs diminish rapidly and the seller needs quickly to achieve scale. For example, initial production batches for some products are so costly that they cannot realistically be sold at a level that recovers even the incremental cost, but costs of the product diminish rapidly. In some cases, moreover, where average avoidable costs exceed the nominal price there may be no anticompetitive effect because the revenues realized in the long run exceed such costs. In the United States subsidizing mobile phone handsets is common and justified, for example, if the future stream of wireless service revenues generated in connection with the sale of handsets is profitable even after taking into account the subsidy for the handset.

E. "Targeted Pricing" in the Telecommunications Industry

Targeted pricing occurs when a company that has sufficient information to identify its customers' habits engages in price discrimination and charges different customers different prices. In general, such price discrimination is procompetitive. For example, in differentiated product industries like telecommunications, targeted pricing can allow companies to expand their product offerings. Targeted pricing may also enhance competition when switching costs prevent customers from trying out new products or switching to products that might increase their utility. Targeted pricing that reduces these switching costs is procompetitive.

While targeted pricing may cause competitors to exit the market or hinder their entry, the same can be said of aggressive price competition. Thus, the problems associated with false positives that arise in the predatory pricing context would be found in any attempt to prosecute targeted pricing, even assuming such conduct could be defined in a tractable manner. Accordingly, the Sections recommend that the Bureau exercise caution in deciding whether to prosecute above-cost targeted pricing. If the Bureau decides to do so, then the Sections recommend that the Bureau carefully define "targeted pricing," establish safe harbors, and delineate a clear test for what specific conduct the Bureau would consider anticompetitive.

If multiple companies engage in targeted pricing, then the practice will decrease average prices throughout the industry and benefit consumers. If the Bureau chooses to prosecute targeted pricing in a particular instance under criteria identified in the Draft Bulletin, then the Sections recommend that the Bureau should first determine: (1) whether the dominant firm has sufficient information to engage in targeted pricing, as the ability to identify customers is a necessary condition for the practice; (2) whether the firm's pricing is a significant deviation from its current pricing structure; (3) what conduct constitutes a significant deviation; (4) whether other firms have the capability to respond to targeted pricing themselves; and (5) if those other firms are unable to respond, whether the dominant firm's targeted pricing that could drive existing firms out of the market or create a barrier to entry. If all these conditions are present, then the Sections recommend the Bureau consider investigating only targeted pricing that is correlated in time with a specific competitive threat, e.g., entry or expansion.²⁵

F. Bundling

The approach to tying and bundling set out in the Draft Bulletin is very constructive. In particular, the Sections commend the Bureau for recognizing that tying and bundling often benefit consumers. The Sections recommend that the Bureau also recognize that the technical integration of two products may constitute a new product, rather than a tie or bundle. The Sections further commend the Bureau for recognizing that bundling sometimes may be viewed as a means to raise rivals' costs or to engage in predation.

The Sections note that the absence of objective justification or efficiencies is one of the elements necessary to establish an abuse. Given the often procompetitive nature of bundling, the Sections recommend that the Bureau clearly indicate that it will challenge the conduct only when there is an absence of objective justifications or efficiencies.

The Bureau may wish to describe further the degree of foreclosure that will give rise to concern, given that the analysis of foreclosure effects in individual cases will be highly fact specific. Use of a single consistent term to describe the level of foreclosure that will give rise to concern (e.g., "substantial part" of the market), and guidance on what level normally will not give rise to concern in the absence of special circumstances would be beneficial. The Sections suggest the Bureau provide further concrete guidance by identifying "safe harbors," which would be especially helpful to businesses trying to evaluate the legality of proposed courses of action.

IV. Substantial Lessening or Prevention of Competition

The Bureau may wish to address whether potential efficiencies present in the telecommunications industry could indicate a lack of anticompetitive intent. Where firms seek to offer "less expensive, more efficient, and [a broader range of] services,"²⁶ the intent, rather than being anticompetitive, may be to improve efficiencies that, in the end, could benefit consumers.

V. Remedies

The Bureau prefers a "remedy agreed upon by the parties affirmed in a consent agreement."²⁷ The Sections commend this approach because negotiation of remedies by the parties reduces the cost and delay associated with litigation and enables the Bureau to meet its concerns of ensuring that competition is not lessened "substantially" without the need for protracted litigation. Negotiated remedies provide better results for both the agencies and the parties.

The Bureau may wish to consider adding language stating that the purpose of remedial action is to promote competition, not protect competitors. The Sections believe that the Bureau should not be concerned with the effect of a transaction on inefficient competitors.

---

¹ The members of the working group that drafted these comments are Bernard Nigro, Joseph Simons, Aryeh Friedman, Julie Soloway, Ankur Kapoor, Mark Katz, Shuli Rodal, Carl Willner, Cynthia Lagdameo, Jonathan Lave, Jose Tesoro, Micah Wood and Richard Elliott.

² Draft Bulletin, Part 2.4, at 7-8.

³ For example, market share is not necessarily an indication of market power for a historic monopoly. A firm that has seen its share drop from 100% to 80% in two years may have no market power at all, let alone a position "approaching monopoly."

⁴ In Europe, for example, "dominance can be presumed in the absence of evidence to the contrary if an undertaking has a market share persistently above 50 percent." Office of Fair Trading, Abuse of a Dominant Position (2004) ¶ 4.18 (footnote omitted), available at http://www.oft.gov.uk/NR/rdonlyres/0620258B-3006-4B1C-ADC65CC69E6EF4F1/ 0/OFT402.pdf.

⁵ Enforcement Guidelines on the Abuse of Dominance Provisions (July 2001) at 2, available at http://www.competitionbureau.gc.ca/epic/site/cb-bc.nsf/en/01251e.html (stating that "a market share of less than 35 percent will generally not give rise to concerns of market power or dominance [while a] market share of more than 35 percent will generally prompt further examination").

⁶ Id. at 16.

⁷ For a thorough explanation of network effects and their impact on competition and consumer value, see Brian A. Facey & Micah Wood, Economic Utilitarianism and Merger Efficiencies in Network Industries, Presented at the 2004 Canadian Bar Association, Annual Fall Conference on Competition Law, Ottawa, Ontario, Sept. 23-24, 2004.

⁸ Draft Bulletin, Part 3.4 at 13

⁹ Draft Bulletin, Part 4.2.1, at 16.

¹⁰ Id.

¹¹ Id.

¹² Id. at 16-17.

¹³ But see Case COMP/C-1/37.451, 37.578, 37.579 Deutsche Telekom AG, (2003), available at http://www.worldlii.org/eu/cases/ECComm/2003/18.html (Commission held that it is sufficient to establish a margin squeeze if there is "a disproportion between the two charges such that competition is restricted"; relevant question was whether the difference between the retail prices charged by the dominant undertaking and the wholesale prices it charged its competitors for comparable services was "negative or insufficient to cover the product specific costs to the dominant operator of providing its own retail services on the downstream market.").

¹⁴ Draft Bulletin, Part 4.2.2, at 17.

¹⁵ Id.

¹⁶ See Enforcement Guidelines on the Abuse of Dominance Provisions – Appendix III: Exclusionary Vertical Squeezing ("In situations where it is established that one supplier (wholesaler) possesses the market power required to exercise control, and that this control has been acquired through means that do not contravene the Act, potential monopoly profits can be extracted simply by charging a monopoly price for the product at the wholesale level. This is not an abuse of market power. In these instances it is not in the supplier's interest to charge a price that would eliminate or discipline his or her customers, as he or she is already extracting the maximum return.").

¹⁷ See id.

¹⁸ See id. ("Clearly, determining when squeezing is anti-competitive involves careful analysis.").

¹⁹ Draft Bulletin, Part 4.2.2, at 17.

²⁰ Phillip Areeda, Essential Facilities: An Epithet in Need of Limiting Principles, 58 ANTITRUST L.J. 841, 852 (1990).

²¹ Draft Bulletin, Part 4.2.2, at 18.

²² Verizon Communications, Inc. v. Law Offices of Curtis V. Trinko, 540 U.S. 398, 415 n.4 (2004). These standards are preferable to those announced by the European Commission. See Case C-418/01, IMS Health GmbH & Co. OHG v NDC Health GmbH & Co. KG (stating that "[a] company which has a dominant position in the provision of facilities which are essential for the supply of goods or services abuses its dominant position where, without objective justification, it refuses access to those facilities"). The European Commission's test is overbroad because it would call into question the "objective justification" of all refusals by a dominant firm to assist its rivals and thereby chill legitimate dominant firm conduct.

²³ See, e.g., Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209 (1993) (applying average variable cost standard, but declining to resolve conflict among circuits as to appropriate measure of cost); Spirit Airlines, Inc. v. Northwest Airlines, Inc., 431 F.3d 917 (6th Cir. 2005) (applying average variable cost standard); Taylor Publ'g Co. v. Jostens, Inc., 216 F.3d 465 (5th Cir. 2000) (same); Irwin Indus. v. Goodyear Aerospace Corp., 974 F.2d 241 (2d Cir. 1992) (same) (citing Phillip Areeda & Donald F. Turner, Predatory Pricing and Related Practices under Section 2 of the Sherman Act, 88 HARV. L. REV. 697, 716-18, 733 (1975). In the United States, the Second, Fifth, Seventh, and District of Columbia Circuits have largely accepted professors Phillip Areeda's and Donald F. Turner's rule that only prices below average variable cost should be presumed unlawful. The Third and Tenth as well now lean toward the average variable cost rule. Three circuits – the Eighth, Ninth, and Eleventh – agree that below average variable cost pricing is presumptively predatory, but they permit the plaintiff to produce evidence of a defendant's predatory intent where prices are above average variable cost but below average total cost. The Fourth Circuit has yet to enunciate which approach it follows. And although they accept that prices below average total cost are lawful and those below average variable cost are unlawful, the First and Sixth Circuits have not clarified their approach to prices that are above average variable cost yet below average total cost.

²⁴ Clearly, a cost standard is crucial to determining that predation has occurred or is occurring. There should be a single cost standard that separates pricing policies that could be predatory from those that clearly are not. According to economic theory, the best standard would be a "proxy" for marginal cost. In most industries, the appropriate proxy would be average variable cost or average avoidable cost, and an economic case may be made that average avoidable cost is a better benchmark than average variable cost. See William J. Baumol, Predation and the Logic of the Average Variable Cost Test, 39 J. LAW & ECON. 49 (1996); Brian A. Facey & Dany H. Assaf, Monopolization and Abuse of Dominance in Canada, the United States and the European Union: A Survey, 70 ANTITRUST L.J. 513, 552-55 (2002).

²⁵ Given that price discrimination is generally welfare enhancing, there would seem to be even less reason to be concerned about targeted pricing as long as it is not predatory. See supra, Part E, at 11.

²⁶ Calvin S. Goldman, Ilene Knable Gotts, and Michael E. Piaskoski, The Role of Efficiencies in Telecommunications Merger Review, 56 FED. COMM. L.J. 93 (2003).

²⁷ Draft Bulletin, Part 6.1, at 23.