For decades, the received wisdom was that merger analysis always started with the definition of relevant product and geographic markets, and that the resolution of any challenged horizontal merger would require structural remedies with, at most, the rarest resort to behavioural remedies for vertical mergers. The 2010 Horizontal Merger Guidelines1 issued by both the DoJ and the FTC, and the more recent 2011 Remedies Guide2 issued by the DoJ, have begun to modify this received wisdom. Recent merger cases, and their resolution, show, however, that market definition still holds its often ‘critical’3 role in merger analysis, especially when cases head to court, even though the Guidelines relegate market definition to a supporting role. In contrast, with respect to the negotiation of remedies, where the agencies are less subject to judicial constraint, in a short time, the agencies have been able to build a respectable record of using behavioural remedies as some or all of the solution for challenged mergers, including horizontal mergers.
Continued importance of market definition
The 2010 Guidelines make clear that the agencies consider market definition to be a means, not an end, and explain that market definition is useful only to the extent that it ‘illuminates’ likely competitive effects. Even though the agencies’ merger analysis may ‘not start with market definition,’4 market definition nonetheless continues to play a significant role in merger analysis under section 7 of the Clayton Act. The reason for this is straightforward: federal district courts continue to define the relevant product market and geographic market as a threshold matter in challenges to proposed mergers under section 7. Moreover, established precedent grants the agencies a presumption of competitive harm in litigated cases upon a showing of ‘undue concentration’ in a well-defined market, and the agencies are understandably reluctant to give up such a litigation advantage.5 In any event, the agencies have not yet been bold enough to ask a district court to stop a merger with a complaint based solely, for example, on direct evidence of unilateral effects or an argument based only on elimination of a ‘maverick’ competitor.6 Accordingly, as a practical matter, market definition remains the threshold issue in horizontal merger analysis.
Two of the headline merger challenges in 2011 are strong evidence of this critical role that market definition continues to play in analysis of proposed mergers under section 7. Market definition was a central issue in the DoJ’s challenge of both H&R Block’s proposed acquisition of TaxACT and AT&T’s proposed acquisition of T-Mobile.7 While each of the H&R Block and AT&T acquisitions present unique market definition issues, these cases are reminders that:
- market definition is not an abstract, theoretical exercise, but instead a process to identify the competitors that in fact do and will constrain the merging parties post-transaction; and
- the perception that each merging party conveys in its internal documents and other communications as to the relevant pool of competitors is very strong evidence as to market definition.
H&R Block’s proposed acquisition of TaxACT
H&R Block and TaxACT both sell what the DoJ came to refer to as ‘digital do-it-yourself tax preparation products’ (DDIY). These products consist of software that assists and guides individuals in preparing tax returns, and are available online or on disks that can be loaded onto the taxpayer’s computer. An analysis of the complaint filed by the DoJ in May 2011 to enjoin the acquisition would not alone suggest that the H&R Block case would be focused upon market definition. The market definition and concentration allegations of the DoJ’s complaint consist of four short paragraphs defining the relevant product market as DDIY products,8 a single paragraph defining the relevant geographic market as national,9 and a cursory treatment in three paragraphs of market concentration.10 In contrast, the DoJ’s complaint contains lengthy allegations concerning TaxACT’s history as a disruptive or ‘maverick’ firm, and an extensive treatment of the risks of unilateral price increases or reduction in innovation arising from the elimination of head-to-head competition.11
Given this relative attention to market definition and concentration issues, it is possible that the DoJ simply did not view H&R Block as a particularly difficult case to win on market definition issues. It is also possible, however, that the DoJ’s complaint was an attempt to move away from the traditional market concentration analysis and to focus on two direct sources of evidence of anti-competitive effects - unilateral effects of eliminating a head-to-head competitor and the potential for coordinated effects from eliminating a ‘maverick’ firm. If the DoJ’s complaint was an effort to move away from the traditional analysis, the district court’s response indicates that the court might not follow the DoJ’s lead any time soon. The district court engaged in the standard Baker Hughes analysis, starting with the determination of product and geographic markets, to which the court dedicated more than half of the 30-page substantive section of its opinion.12 The court then analysed concentration levels in the market and concluded that the DoJ was entitled to a presumption that the merger would lessen competition and, thus, had established a prima facie violation of section 7. As a result, the DoJ’s theories of competitive harm were not tested directly. Instead, they were discussed only obliquely, in the context of the defendants’ efforts to rebut the presumption of both coordinated and unilateral effects arising from concentration levels in the defined DDIY market.13
The market definition issue in H&R Block can be stated simply: whether other methods of tax preparation should be included with DDIY in the product market for purposes of merger analysis. In addition to DDIY, there are two other methods of tax preparation: ‘pen-and-paper’, which involves manual preparation of printed tax forms or completion of electronic forms on the Internal Revenue Service’s website; and ‘assisted preparation’, which involves hiring a tax professional to prepare the tax returns. The merging parties, of course, argued that the relevant product market included all these tax preparation methods, while the DoJ argued for the narrower product market of only DDIY. The district court ultimately agreed with the DoJ.
The trial included ‘duelling experts’ and the court eventually analysed the experts’ testimony and data at length, but relied on none of the experts’ analyses. Instead, the court relied upon traditional evidence - the parties’ own, sometimes damning, documents and the kind of practical indicia endorsed in Brown Shoe14 - to conclude that DDIY is the relevant product market.
The district court reviewed at length the merging parties’ perception of the market - as reflected in internal documents and testimony. The parties’ own words in these documents and testimony demonstrated that only DDIY - and neither pen-and-paper nor assisted preparation - exerted a competitive influence on the merging parties. The court was persuaded by what it characterised as ‘strong evidence’:
- that TaxACT tracked marketing, product offerings and pricing of only DDIY competitors;
- that TaxACT determined its pricing and business strategy only in relation to DDIY competitors;
- that investment banker-prepared documents identified DDIY competitors as the ‘primary’, ‘major’ or ‘main’ competitors;
- that H&R Block consultant-prepared documents tracked the activities, prices and product offerings of only DDIY competitors;
- that analysis of the proposed transaction measured market share in a DDIY market; and
- that H&R Block personnel conceded in testimony that prices and activity in the DDIY business did not impact H&R Block’s assisted preparation business.
To bolster its heavy reliance on this evidence, the district court cited Whole Foods15 for the proposition that ‘economic actors usually have accurate perceptions of economic realities.’16
The court also relied heavily on the distinctive attributes of the products (including differing user experiences) and differences in those attributes that make DDIY distinct from other methods of tax preparation. In particular, the district court explained that DDIY products involve ‘different technology, price, convenience level, time investment, mental effort and type of interaction’ than other tax preparation methods. It is fair to say that, having analysed the differences between DDIY and the other tax preparation methods - including product characteristics and substantially different pricing - the court did not find defendants’ all-inclusive proposed market definition very plausible at all. Indeed, the court cites Whole Foods for the analogy that automobiles do not compete with horse-drawn carriages.
Having rejected the defendants’ broad market definition, the district court then spends several pages of its opinion on an after-the-fact examination of expert testimony.17 The district court ultimately concluded that the expert testimony was not particularly helpful but, on balance, supported the conclusion that the relevant product market was DDIY.18
The H&R Block district court’s approach to market definition is certainly not new. In 2009, for example, the court in FTC v CCC Holdings Inc19 encountered very similar circumstances. The merging parties were two firms that sold ‘total loss software products’ (TLV), which allow automobile insurers to value an automobile deemed a total loss. The merging parties argued that the TLV products were in the same market as various hard-copy books that report automobile values (the Books). After examining expert testimony, the court explained that, ‘[n]otwithstanding’ the testimony, ‘real-world evidence shows that Books and TLV are not part of the same product market,’ and cited Arch Coal for the proposition that ‘antitrust theory and speculation cannot trump facts.’20 CCC Holdings court’s ‘real-world’ analysis was the same as the ‘practical’ analysis performed by the district court in H&R Block.
H&R Block and CCC Holdings stand as important reminders that ‘practical’ and ‘real-world’ evidence can be more important than expert testimony and economic theory when it comes to market definition. Since the point of market definition is to identify the pool of competitors that meaningfully constrain the merging parties, evidence that each merging party’s business decisions are influenced by only a subset of all potential competitors is compelling evidence that only that set of competitors should be included in the relevant market. Even if a broader set of firms might compete with the merging parties ‘at some level,’ that set of competitors is not necessarily included for purposes of merger analysis.21
AT&T’s proposed acquisition of T-Mobile
The DoJ’s highly publicised challenge in September 2011 of AT&T’s proposed acquisition of T-Mobile, one of only three other nationwide mobile wireless telecommunications services providers, is yet another illustration of market definition’s key role in identifying the competitors that will, in fact, constrain the merging parties post-transaction. The AT&T market definition analysis is particularly interesting because it approaches the definition of geographic markets in a somewhat unusual manner, but one that the DoJ and the Federal Communications Commission (FCC) would suggest reflect real-world competitive dynamics.22 The DoJ essentially alleged one product and related geographic market for ‘enterprise and government’ customers and another for everyone else, ie, retail customers. Although the DoJ approached these markets in very different ways, the result was the same - the defined markets included only the four nationwide competitors.
The DoJ’s treatment of business and government customers as a distinct product market with a national geographic scope is not surprising. The DoJ appeared to be taking a Brown Shoe approach and basing the allegation of the separate market on differences in needs and approaches to buying the services. The DoJ alleged that enterprise and government customers needed providers who offered nationwide service that would cover an enterprise’s employees who were disbursed throughout, or needed to travel throughout, the United States. The DoJ also alleged that business and government customers bought these nationwide services through a bidding process that was distinctly different from the buying process of retail customers, who acquired services through retail outlets. The DoJ contended that the geographic market for these enterprise and government customers was national in scope because these customers would deal with a provider anywhere in the US so long as the provider had a national network.
The market definition analysis for retail customers was a different and more interesting matter, however. The DoJ again alleged a product market of mobile wireless telecommunications services,23 but had a lengthy discussion of the relevant geographic market in which it alleged that ‘local areas’, described as FCC-defined Cellular Market Areas (CMAs), constituted relevant geographic markets for retail consumers who acquire such services from providers’ outlets near where they work or live.24 The DoJ, however, was very specific in limiting the scope of relevant competition in these local areas to only ‘national competition [...] conducted in local markets’.
At first blush, one is tempted to think that this ‘neither fish nor fowl’ market definition was designed by a committee that could not make up its mind. Instead, it is likely just the DoJ’s way of expressing their view of the expected outcome of the Guidelines’ hypothetical monopolist test applied to any given ‘local market’. In other words, the DoJ apparently believed that, if a hypothetical monopolist owned or controlled all four of the nationwide providers and imposed a price increase, that increase would be profitable because too few customers in the local market would be willing to switch from nationwide suppliers to a regional or local supplier. Indeed, the FCC’s Staff Analysis suggests that the FCC, and perhaps the DoJ as well, believed that, because other non-nationwide competitors at most covered 34 per cent of the US population, a price increase by the hypothetical monopolist would be profitable across the country, because too few retail customers would switch to regional or local providers across all the local markets.25
In the end, the DoJ gets to the same place with retail customers as with business and government customers - only four competitors in the market. The route to get there is a strange one, however. The DoJ essentially simultaneously defines the geographic market from both the retail customers’ perspective (local) and the suppliers’ perspective (nationwide)26 and calls it ‘nationwide competition [...] conducted in local markets’.
As is often the case, AT&T’s documents and statements also played a role in supporting the DoJ’s ‘neither fish nor fowl’ market theory. According to the complaint, AT&T had stated in a recent merger review that AT&T develops its pricing and other plans on a national basis and that ‘the predominant forces driving competition among wireless carriers operate at the national level.’ In particular, the DoJ quotes an AT&T document in which AT&T states that it sets its ‘rate plans, features and prices’ in response to national providers.27 AT&T was reportedly also contending that the analysis should be conducted on local markets. The DoJ combined the two admissions to arrive at its unusual ‘national competition [...] in local markets’ geographic market definition.
One wonders whether the DoJ might have reached the same place by distinguishing between the service provided by nationwide competitors and that provided by regional or local service competitors in defining the product market. Local providers must provide coverage outside their local areas through roaming arrangements that provide a different customer experience due to roaming fees, and the DoJ did allege that the difference in bidding was sufficient to differentiate business and government customers from retail customers. In addition, the DoJ alleges that national providers’ nationally recognised brands, reputation and access to the most popular handsets, such as the iPhone, could also make local providers’ service distinguishable from those of regional providers.
The DoJ was probably concerned about remaining consistent with product market treatment of past mobile telecommunications services mergers by both the DoJ and the FCC. The distinctions between nationwide and regional service may also have seemed too small in comparison, for example, to those in H&R Block and CCC Holdings - hard-copy products versus electronic products. However, there is a precedent for making this kind of product distinction. As the court in FTC v Staples, Inc explained, ‘the mere fact that a firm may be termed a competitor in the overall marketplace does not necessarily require that it be included in the relevant product market for antitrust purposes.’ In Staples, the court defined the relevant product market as the ‘sale of consumable office supplies through office supply superstores’, and excluded from the relevant product market the sale of consumable office supplies through the mail and by stores such as Wal-Mart. The court justified the distinction based largely on evidence that the office supply superstores reacted competitively only to other office supply superstores. Accordingly, notwithstanding the fact that the physical products sold were frequently identical, the court looked at ‘real-world’ evidence to define a narrower pool of competitors that meaningfully constrained office supply superstores. Utilising the H&R Block roadmap, and Staples as precedent, it is plausible to conclude that AT&T could have been framed as a product market of nationwide suppliers. Indeed, the FCC suggested just that and described the alternative market as a ‘national market for retail wireless
Conduct remedies become more common
The DoJ’s June 2011 revisions to its Policy Guide to Merger Remedies formally recognises for the first time that conduct remedies can be an effective means of eliminating harm to competition, while ‘preserving the beneficial aspects of the merger’. The 2011 Remedies Guide does explain that conduct remedies are more likely to be effective in addressing concerns raised by vertical integration, but that conduct remedies are ‘sometimes’ useful, especially in conjunction with a structural remedy, to address competitive concerns raised by horizontal mergers. The 2011 Remedies Guide provides a non-exhaustive list of potential conduct remedies that could be implemented in a merger consent decree, including:
- firewall requirements;
- prohibitions on customer discrimination;
- mandatory licensing provisions;
- transparency requirements;
- anti-retaliation provisions; and
- prohibitions on certain contracting practices.
The 2011 Remedies Guide emphasises the still challenging feature of conduct remedies: the need for continuous monitoring to insure adherence to the conduct requirements. The 2011 Remedies Guide identifies two basic enforcement mechanisms: monitors and arbitration. The DoJ can appoint a monitor to whom the relevant party submits periodic reports. Alternatively, the consent decree can enable affected parties to enforce the conduct remedies through arbitration.
The range of possible conduct solutions is shown by the consent decree that Google, Inc entered into in connection with its acquisition of ITA Software, Inc (ITA), which may well have been a test of the concepts set forth in the Merger Remedies Guide about a year later.29 ITA Software, Inc develops and licenses QPX, software that provides flight search functionality for airlines, online travel agents and online travel search sites. After acquiring ITA, Google planned to offer an online travel search product that would compete with existing travel sites, but the DoJ was concerned that Google might harm competition by restricting access to QPX or permitting QPX’s quality or ongoing innovation to decline. To preclude this, the final judgment contained a number of conduct remedies:
- Mandatory licensing - Google is required to honor existing QPX licences for existing travel search sites, renew existing licences for up to five years, and offer licences to other travel sites not currently under contract. Comparable mandatory licensing requirements also apply to a new product in development.
- Restriction on contract practices - Google is restricted in its ability to prohibit licencees from using alternative products. Google is also prohibited from entering into agreements with airlines that would restrict an airline’s right to share information with Google’s competitors.
- Research and development commitment - to avoid Google’s permitting QPX to deteriorate, Google is required to continue improving the QPX product. In particular, Google must ‘devote substantially the same resources to research, development and maintenance of QPX [...] as ITA did in the average of the two years prior.’ Google is also required to complete development of a new product.
- Non-discrimination - Google must offer ordinary course upgrades to QPX to its online travel search competitors at the same price that the upgrades are made available to other customers.
- Firewall - Google is required to establish a firewall that will prevent the misuse of competitively sensitive information received from online travel search competitors in the course of providing the QPX product to those competitors.
To enforce the first four conduct remedies, the DoJ is relying upon enforcement by competitors and other affected parties pursuant to an arbitration provision. While the final judgment does not appoint a monitor to oversee compliance, it does grant the DoJ access to Google’s records and documents to ensure compliance, if necessary.
Most of the conduct remedies, such as mandatory licensing or firewall maintenance, in Google/ITA are not terribly unusual. The requirement that Google continue to develop and innovate with respect to travel search software is unusual, however. The final judgment in Google/ITA requires Google both to continue to invest in development of QPX, and to continue to invest in a not yet commercially available product. This remedy is unusual because it appears that it would be hard to enforce and monitor. While the consent decree contains specific expenditure requirements with respect to development efforts, time will tell whether this unusual remedy works.
Since the release of the Remedies Guide, the DoJ has suggested that Google/ITA was not entirely unique by entering into a consent decree in a horizontal merger with a requirement similar to Google’s ‘development’ requirement. In the acquisition by George’s, Inc (George’s) of a chicken processing complex in the Shenandoah Valley from Tyson Foods, Inc (Tyson),30 the DoJ was concerned that, after the acquisition, George’s would exercise monopsony power and reduce the output produced in the acquired facility. The final judgment required George’s to make a number of investments in and improvements to the acquired processing facilities on the theory that these investments would force George’s to run the plant at capacity in order to recover the investments. The DoJ believed that running the plant at capacity would have a pro-competitive effect. This is clearly an innovative remedy that is definitely a leap beyond traditional structural remedies for horizontal mergers.
Clearly the FTC, and the DoJ in particular, have been engaged in substantial innovation in merger analysis in recent years. That innovation has been most evident in the merger remedies being devised in consent decree negotiations and less obvious in its substantive approach to mergers that head to the courthouse.
- The US Department of Justice (DoJ) and the Federal Trade Commission (FTC) Horizontal Merger Guidelines issued 19 August 2010 (http://www.justice.gov/atr/public/guidelines/hmg-2010.html) (Guidelines).
- Antitrust Division Policy Guide to Merger Remedies at http://www.justice.gov/atr/public/guidelines/272350.pdf.
- See US v H&R Block, Inc, 2011-2 Trade Cas. paragraph 77,678 at 122,011 (DDC 2011).
- Guidelines at 7.
- US v Baker Hughes Inc, 908 F2d 981 (DC Cir 1990); FTC v CCC Holdings Inc, 605 F Supp 2d 26, at 36 (DDC 2009).
- Guidelines at 3, 20.
- H&R Block was acquiring 2SS Holdings, Inc, which, in turn, owned the TaxACT products.
- Complaint in US v H&R Block, Inc, et al available at http://www.justice.gov/atr/cases/f271500/271579.html (H&R Block Complaint) paragraphs 23-26.
- H&R Block Complaint paragraph 27.
- H&R Block Complaint paragraphs 37-39.
- H&R Block Complaint paragraphs 28-36, 40-49.
- H&R Block at 122,011.
- Indeed, when the court discussed the DoJ’s argument concerning TaxACT’s status as a ‘maverick firm, the court even called it a game of ‘semantic gotcha’ over which the parties ‘spilled substantial ink’ and complained that the ‘maverick’ or ‘disruptive’ firm argument lacked a ‘clear standard, based on functional or economic considerations.’ H&R Block at 122,029. It should be noted that, in the course of finding defendants had not rebutted the presumption of coordinated effects, the court nonetheless observed that TaxACT was consistently the primary ‘maverick’ as to pricing, albeit not as to innovation.
- Brown Shoe Co v US, 370 US 294 (1962).
- FTC v Whole Foods Market, Inc, 548 F3d 1028 (DC Cir 2008).
- H&R Block at 122,013.
- H&R Block at 122,014.
- H&R Block at 122,014. It should be noted that Judge Howell’s opinion was not an indictment of economic analysis in general. The strength of the economic work in this case was undermined by the lack of any suitable data for the economists to use.
- FTC v CCC Holdings, Inc, 605 F Supp 2d 26 (DDC 2009).
- FTC v Arch Coal, Inc, 329 F Supp 2d 109 (DDC 2004)
- See H&R Block at 122,014.
- Because AT&T and T-Mobile abandoned their merger, the only detailed factual statements available from the DoJ are those in the complaint. Because the FCC also had to approve the merger, the FCC did, however, issue a lengthy report that details relevant facts that appear consistent with the allegations of the DoJ’s complaint. See FCC, Staff Analysis and Findings, WT Docket No. 11-65 (FCC Staff Analysis).
- The main issue on product market definition for retail customers was the fairly obvious point that landline telecommunication services were not a substitute for mobile wireless services. See AT&T Complaint paragraph 12.
- AT&T Complaint paragraph 17.
- FCC Staff Analysis paragraphs 34 and 38.
- Paragraph 20 of the AT&T Complaint seems to acknowledge this fact by stating: ‘Whereas CMAs are appropriate geographic markets from the perspective of individual consumer choice, from a seller’s perspective, the Big Four [nationwide] carriers compete against each other on a nationwide basis and AT&T’s acquisition of T-Mobile will have nationwide competitive effects across local markets.’
- AT&T Complaint paragraphs 19-20.
- FCC Staff Analysis at n109.
- Final Judgment in US v Google Inc and ITA Software, Inc at http://www.justice.gov/atr/cases/f275800/275897.pdf.
- Final Judgment and Competitive Impact Statement in US v George’s Foods LLC available at http://www.justice.gov/atr/cases/f278500/278560.pdf and http://www.justice.gov/atr/cases/f272500/272501.pdf.