UNITED STATES OF AMERICA v. H&R BLOCK, INC. et al, No. 1:2011cv00948 - Document 108 (D.D.C. 2011)

Court Description: MEMORANDUM OPINION. This is the public version of the Memorandum Opinion filed under seal on October 31, 2011. Signed by Judge Beryl A. Howell on 11/10/2011. (lcbah2) (Entered: 11/10/2011)

Download PDF
UNITED STATES OF AMERICA v. H&R BLOCK, INC. et al Doc. 108 UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA UNITED STATES OF AMERICA, Plaintiff, v. Civil Action No. 11-00948 (BAH) H&R BLOCK, INC., et al., Defendants. MEMORANDUM OPINION1 Last year, approximately 140 million Americans filed tax returns with the Internal Revenue Service (“IRS”). Paying taxes is a fundamental civic duty in our democracy. Taxes pay for the government to carry out its constitutionally mandated functions and enable the government to give force to the laws and policies adopted by the people of the United States through their elected representatives. Despite the necessity of taxes to fund our government and to sustain services that many citizens depend upon, the task of preparing a tax return brings joy to the hearts of few. Many find it to be a complex and tedious exercise. Fortunately, various businesses offer different products and services designed to assist taxpayers with preparing their returns. These tax preparation businesses principally include accountants, retail tax stores, and digital tax software providers – all of which provide important services to the American taxpayer. In this case, the United States, through the Antitrust Division of the Department of Justice, seeks to enjoin a proposed merger between two companies that offer tax software 1 The Court provided this Memorandum Opinion to the parties in final form on October 31, 2011, but public release was delayed to ensure that no confidential business information that had been submitted under seal was released. Based on input from the parties, confidential business information has been redacted from the opinion, with such redactions reflected by the insertion of the text “{redacted}.” In some instances, redacted confidential business information has been replaced by more general language that reflects the same underlying concepts without revealing the confidential business information. Such substitutions are indicated by braces surrounding the substituted text. Dockets.Justia.com products – H&R Block and TaxACT – on the grounds that the merger violates the antitrust laws and will lead to an anticompetitive duopoly in which the only substantial providers of digital tax software in the marketplace would be H&R Block and Intuit, the maker of the popular “TurboTax” software program. After carefully considering all of the evidence, including documents and factual and expert testimony, the applicable law, and the arguments before the Court, the Court will enjoin the proposed merger for the reasons explained in detail below. * * * 2 TABLE OF CONTENTS I. BACKGROUND………………………………………………………………... 4 A. Overview……………………………………………………………………... 4 B. The Merging Parties……………………………………………………......... 7 C. The History Of TaxACT And The Proposed Transaction……………….. 8 D. Free Products And The Free File Alliance……………………..………….. 10 II. STANDARD OF REVIEW………………………………………………........ 13 III. DISCUSSION……………………………………………………………........ 15 A. The Relevant Product Market………………………………………........... 15 1. The Defendants’ Documents Show That DDIY Is The Relevant Product Market……………………………………………………………………. 19 2. The Relevant Product Market Does Not Include Assisted Tax Preparation Or Manual Preparation…………………………………………….. 22 3. The Economic Expert Testimony Tends To Confirm That DDIY Is The Relevant Product Market……………………………………………………. 31 B. Likely Effect on Competition………………………………………………. 50 1. The Plaintiff’s Prima Facie Case……………………………………........ 50 2. Defendants’ Rebuttal Arguments………………………………………... 53 a. Barriers to Entry………………………………………………............ 53 b. Coordinated Effects………………………………………………….. 60 c. Unilateral Effects… …………………………………………….......... 67 d. Post-Merger Efficiencies….…………………………………….......... 80 IV. CONCLUSION………………………………………………………………. 86 3 I. BACKGROUND A. Overview The United States, through the Antitrust Division of the Department of Justice (the “DOJ,” the “government,” or the “plaintiff”), filed this action on May 23, 2011. The DOJ seeks to enjoin Defendant H&R Block, Inc. from acquiring Defendant 2SS Holdings, Inc. (“TaxACT”), which sells digital do-it-yourself tax preparation products marketed under the brand name TaxACT. Compl. ¶ 10. H&R Block (“HRB”) is a Missouri corporation headquartered in Kansas City, Missouri. Id. ¶ 9. 2SS Holdings, or TaxACT, is a Delaware corporation headquartered in Cedar Rapids, Iowa. Id. ¶ 10. Defendant TA IX, L.P. (“TA”), a Delaware limited partnership headquartered in Boston, Massachusetts, owns a two-thirds interest in TaxACT. Id. ¶ 11. As noted above, approximately 140 million Americans filed tax returns with the IRS in 2010. Id. ¶ 1. Broadly speaking, there are three methods for preparing a tax return. The “pen and paper” or “manual” method includes preparation by hand and with free, electronically fillable forms available on the IRS website. A second method, known as “assisted” preparation, involves hiring a tax professional – typically either a certified public accountant (“CPA”) or a specialist at a retail tax store. HRB operates the largest retail tax store chain in the United States. Cobb, TT, 9/19/11 a.m., at 37. The companies Jackson-Hewitt and Liberty Tax Service also operate well-known retail tax stores. Finally, many taxpayers now prepare their returns using digital do-it-yourself tax preparation products (“DDIY”), such as the popular software product “TurboTax.” DDIY preparation is becoming increasingly popular and an estimated 35 to 40 million taxpayers used DDIY in 2010. GX 19 at 3; see also GX 27.2 2 In this opinion, the Court will use the abbreviations “GX”, “GTX”, “DX”, and “DTX” to refer to the government’s exhibits, the government’s trial exhibits, the defendants’ exhibits, and the defendants’ trial exhibits, respectively. 4 The three most popular DDIY providers are HRB, TaxACT, and Intuit, the maker of TurboTax. According to IRS data, these three firms accounted for approximately 90 percent of the DDIY-prepared federal returns filed in tax season 2010.3 GX 27. The next largest firm is TaxHawk, also known as FreeTaxUSA, with 3.2 percent market share, followed by TaxSlayer, with 2.7 percent. Id. The remainder of the market is divided among numerous smaller firms. Id. Intuit accounted for 62.2 percent of DDIY returns, HRB for 15.6 percent, and TaxACT for 12.8 percent. Id. DDIY products are offered to consumers through three channels: (1) online through an internet browser; (2) personal computer software downloaded from a website; and (3) personal computer software installed from a disk, which is either sent directly to the consumer or purchased by the consumer from a third-party retailer. GX 629 at 11. In industry parlance, DDIY products provided through an internet browser are called “online” products, while software applications downloaded onto the user’s computer via the web or installed from a disk are referred to as “software” products. See id. The proposed acquisition challenged in this case would combine HRB and TaxACT, the second and third most popular providers of DDIY products, respectively. According to the government, this combination would result in an effective duopoly between HRB and Intuit in the DDIY market, in which the next nearest competitor will have an approximately 3 percent market share, and most other competitors will have less than a 1 percent share. GX 27. The government also alleges that unilateral anticompetitive effects would result from the elimination of head-to-head competition between the merging parties. Compl. ¶ 45. “TT” refers to trial testimony. “PFF” refers the plaintiff’s proposed findings of fact. “DFF” refers to the defendants’ proposed findings of fact. 3 The denomination of different years in the tax industry can be somewhat confusing. Tax returns are typically due in the month of April following the relevant tax year. Thus, each “tax season” refers to the period when returns for the prior “tax year” are generally completed. For example, “tax season 2010” refers to returns filed primarily in early 2010, corresponding to income earned in “tax year 2009.” 5 Thus, the DOJ alleges that because the proposed acquisition would reduce competition in the DDIY industry by eliminating head-to-head competition between the merging parties and by making anticompetitive coordination between the two major remaining market participants substantially more likely, the proposed acquisition violates Section 7 of the Clayton Act, 15 U.S.C. § 18. Id. ¶¶ 40-49. Accordingly, the government seeks a permanent injunction blocking HRB from acquiring TaxACT. Id. ¶¶ 53-55. On July 6, 2011, the Court entered a scheduling order in this case that provided for an expedited schedule of fact and expert discovery and briefing on the government’s anticipated motion to enjoin the transaction. Joint Scheduling and Case Mgmt. Order, ECF No. 30. On August 1, 2011, the DOJ filed a motion for preliminary injunction against the merger, which was fully briefed by August 18, 2011. The parties subsequently agreed to forego the preliminary injunction phase and proceed directly to a trial on the merits of this action. TT, 9/6/11 a.m., at 89. On September 2, 2011, the Court held a pre-trial conference. On September 6, the Court began a nine-day bench trial that was held on September 6, 7, 8, 9, 12, 13, 15, 19, and 20. Eight fact witnesses and three expert witnesses testified at the hearing. The parties presented testimony from additional witnesses by affidavit and deposition. Each side submitted over 800 exhibits, totaling many thousands of pages. Following the conclusion of the evidentiary phase of the trial, the Court gave the parties approximately two weeks to submit post-trial memoranda and proposed findings of fact, which were filed on September 28, 2011. ECF Nos. 98-99. The Court then heard closing arguments on October 3, 2011. The government’s motion to enjoin HRB’s acquisition of TaxACT is presently before the Court. For the reasons explained in this opinion, the Court grants the government’s motion. 6 Before proceeding to a discussion of the relevant legal standards governing this case, the Court will provide additional background regarding the parties, their proposed transaction, and the tax preparation industry in general. B. The Merging Parties HRB is a Missouri corporation with its principal place of business in Kansas City, Missouri. Compl. ¶ 9; Defs.’ Answer, ECF No. 31, ¶ 9. HRB provides both assisted tax preparation services and DDIY products through separate business units. Bennett, TT, 9/6/11 a.m., at 106. HRB offers its DDIY products for consumers under the brand name “H&R Block At Home” (formerly known as “TaxCut”). GX 629 at 9. In 2011, HRB’s DDIY products generated {significant} revenue. GX 296-2. For the same period, HRB sold approximately 6.69 million DDIY units to consumers. GX 296-2. Separately, in 2011, HRB’s assisted tax preparation business generated approximately $2.7 billion in revenue (based on 14,756,000 U.S. tax returns at an average fee of $182.96, as reported in HRB’s 2011 Annual Report). GX 532 (Cobb Dep.) at 32; GX 565 at 19. 2SS Holdings, Inc. (“2SS”) is a Delaware corporation with its principal place of business in Cedar Rapids, Iowa. Compl. ¶ 10; Defs.’ Answer ¶ 10. 2SS owns 2nd Story Software, Inc., which offers DDIY products under the brand name “TaxACT.” GX 629 at 8-9. In the fiscal year ending April 30, 2011, TaxACT products generated approximately {half as much revenue as H&R Block}. GX 151 at 6. In the same year, consumers used TaxACT to electronically file approximately 5 million federal tax returns. GX 151 at 3-4. TA IX, L.P. (“TA”) is a private equity firm organized under the laws of Delaware with its headquarters in Boston, Massachusetts. Compl. ¶ 11; Defs.’ Answer ¶ 11. In December of 2004, TA purchased a majority interest in 2SS for $85 million, and as a result TA has majority 7 control of 2SS Holdings and 2nd Story Software. GX 55 (Greif Dep.) at 72-73; GX 28-3. C. The History Of TaxACT And The Proposed Transaction TaxACT was founded in 1998 by Lance Dunn and three others, with Mr. Dunn serving as president. Dunn, TT, 9/7/11 p.m., at 49-52. Before founding TaxACT, Mr. Dunn and the other co-founders of the company had worked at Parsons Technology, a software company that had created a DDIY tax preparation product called “Personal Tax Edge.” Id. at 49-52. In 1994, Intuit acquired Parsons Technology and continued to operate Personal Tax Edge as a separate product for approximately two years before merging it into its TurboTax product line. Id. at 51. Mr. Dunn testified that the business objective of founding TaxACT was “to make money selling value tax software which . . . was a category that did not exist at that time” because Intuit’s acquisition of Parsons Technology had eliminated Personal Tax Edge, which had previously occupied a value tax software niche. Id. at 52. Thus, TaxACT “recreated” the category or “niche that the Personal Tax Edge product line filled when it existed.” Id. Over the years, TaxACT has emphasized high-quality free product offerings as part of its business strategy. Id. at 53. TaxACT initially offered a DDIY tax preparation product that made it free to prepare and print a federal tax return, but TaxACT charged a fee for electronic filing (“e-filing”) or preparation of a state tax return. Id. at 54. Thus, from the beginning, TaxACT’s business strategy relied on promoting “free” or “freemium” products, in which a basic part of the service is offered for free and add-ons and extra features are sold for a price.4 As Mr. Dunn put it, “Free is an integral part of the value model. And the beauty of it is it has universal appeal. Everybody likes something for free.” Id. 4 The business model of offering free products and then soliciting customers to purchase additional, related features or services is sometimes referred to as “freemium.” See GX 130 (“H&R Block Strategic Planning Working Session, April 16 &17, 2010”) at 103 (“‘Freemium’ is a known market dynamic that has arisen in multiple product categories and will continue to grow.”). 8 Currently, TaxACT’s free product offering allows customers to prepare, print, and e-file a federal tax return completely for free. Id. at 54; GX 28-10 at 5-7. TaxACT’s “Deluxe” edition, which costs $9.95, contains additional features, such as the ability to import data from a return filed the prior year through TaxACT. GX 55-26; Dunn, TT, 9/7/11 p.m., at 91-92; GX 28-10 at 5-7; GX 28 (Dunn. Dep.) at 219. Customers who use TaxACT to prepare a state tax return in addition to a federal return pay either $14.95 for the state return in combination with the free federal product or $17.95 for the state return in combination with the “Deluxe” federal product. GX 55-26; Dunn, TT, 9/8/11 a.m., at 49. TaxACT’s prices have generally remained unchanged for the past decade. Dunn, TT, 9/7/11 p.m., at 91. The parties first began discussing the potential acquisition of TaxACT by HRB in July 2009. Bowen, TT, 9/15/11 p.m., at 14. During the fall of 2009, teams from HRB and TaxACT met to discuss the possibilities for the potential acquisition and HRB performed due diligence on TaxACT. See DX 244 at 8-9; Bowen, TT, 9/15/11 p.m., at 19-23, 26; DX 9527 at 35. Negotiations between the parties stalled in December 2009 and the proposed deal collapsed. Bowen, TT, 9/15/11 p.m., at 33. The CEOs of the two companies continued to discuss a potential acquisition through the spring of 2010, however. Id. at 34. Serious merger talks resumed in July 2010. Id. at 38-39; DX1005. In October 2010, the HRB Board of Directors approved a plan for HRB to acquire TaxACT. DX 600 at 12-13; Bowen, TT, 9/15/11 p.m., at 59-60. On October 13, 2010, HRB entered into a merger agreement with 2SS and TA. GX 120 at 1. Under this agreement, HRB would acquire control of 2SS for $287.5 million. GX 120 at 6; GX 119 at 1. HRB’s stated postmerger plan is to maintain both the HRB and TaxACT brands – with the HRB-brand focusing on higher priced-products and the TaxACT brand focusing on the lower-priced products. See 9 Bennett, TT, 9/6/11 a.m., at 101-102; DX 1005 at 1. HRB plans {redacted} ultimately to rely on TaxACT’s current technological platform and intends to give Mr. Dunn responsibility for running the combined firm’s entire DDIY business operation from Cedar Rapids, Iowa. Dunn, TT, 9/8/11 p.m. (sealed), at 14-16; see also Bennett, TT, 9/6/11 a.m., at 110. D. Free Products And The Free File Alliance The evolution of TaxACT’s free product offerings and the other free offerings in the DDIY market is important for understanding the claims in this case. The players in the DDIY market offer various “free” tax preparation products, but the features and functionality offered in these free products vary significantly, as do the ways in which these free products are ultimately combined with paid products to earn revenue. While the availability of some types of free product offers has long been a feature of the DDIY market, a spike in free offerings occurred during the last decade in parallel with the growth of e-filing. As a matter of public policy, the IRS actively promotes e-filing because it has an interest in efficient and accessible tax return preparation and filing. The Internal Revenue Service Restructuring and Reform Act of 1998 set a goal of having eighty percent of individual taxpayers e-filing their returns by 2007. IRS Stip., ECF No. 80, ¶ 2. The IRS is close to achieving that goal and the IRS Oversight Board has recommended that the 80 percent benchmark be achieved by 2012. Id. According to stipulated facts attested to by IRS employees, in 2001, the IRS adopted an initiative “to decrease the tax preparation and filing burden of wage earners by providing greater access to free online tax preparation and filing options for a significant number of taxpayers.” Id. ¶ 4. The IRS also determined that it could save a substantial amount of public money by encouraging filers to switch to e-filing, since e-filed returns are cheaper for the IRS to process. Id. ¶ 5. 10 The IRS determined that the most effective and efficient way to accomplish its goal of promoting access to free online tax preparation and filing options was to partner with a consortium of companies in the electronic tax preparation and filing industry. Id. ¶ 6; GX 297D7 at E-2. In 2002, this consortium of companies formed Free File Alliance, LLC (“FFA”) in order to partner with the IRS on this initiative to promote free filing. IRS Stip. ¶ 6; GX 297-D7 at E-2. HRB, TaxACT, and Intuit are all members of the FFA, as are approximately fifteen smaller companies. See IRS Stip. ¶ 8; DX 328. On October 30, 2002, the IRS and the FFA entered into a “Free On-Line Electronic Tax Filing Agreement” to provide free online tax return preparation and filing to individual taxpayers. IRS Stip. ¶ 9. Pursuant to this agreement, members of the FFA would offer free, online tax preparation and filing services to taxpayers, and the IRS would provide taxpayers with links to those free services through a web page, hosted at irs.gov and accessible through another government website. Id. ¶ 12. HRB, TaxACT, and Intuit were among the original members to make free offers through the FFA. Id. ¶ 8. “In 2003, the first year in which free services were available to taxpayers through the FFA, none of the FFA members offered free services to all taxpayers.” Id. ¶ 14. Rather, each “member set eligibility criteria. Most members, including H&R Block, TaxACT, and Intuit, used adjusted gross income (‘AGI’) as a way to define which taxpayers were eligible” for their offers of free federal tax return preparation services. Id. “For example, H&R Block offered free services to taxpayers with an AGI of $28,000 or less.” Id. Some members that offered free federal return preparation services based on AGI also offered free services to taxpayers who met other conditions, such as eligibility to file a Form 1040EZ. Id. “Several members did not define eligibility based on AGI. Of the eleven FFA members that offered free services based on AGI, only TaxACT’s AGI-based offering was available to individuals with AGI over $33,000.” Id. 11 Specifically, TaxACT made its free federal services available exclusively to taxpayers who had AGI over $100,000 or were eligible to file a Form 1040EZ. Id. In 2004, the second year in which free services for federal returns were available to taxpayers through the FFA, TaxACT introduced a new offer through the FFA that offered free preparation and e-filing of federal returns for all taxpayers regardless of AGI or other limitations (“free for all”). See id. ¶ 15; Dunn, TT, 9/7/11 p.m., at 65, 78. After TaxACT introduced a freefor-all offer through the FFA, other companies followed by introducing federal free-for-all offers of their own. Dunn, TT, 9/7/11 p.m., at 78 (“After we offered free for everyone in 2003, in 2004, a lot of companies offered free for everyone on the FFA.”). According to Mr. Dunn’s testimony, after TaxACT made its FFA offer of a free federal product for all taxpayers, without any AGI or other limitations, other companies made efforts to restrict the wide availability of free offers on the FFA. Id. at 79. Specifically, according to Mr. Dunn, Intuit proposed that companies in the FFA collude by agreeing to restrict free offers. Id. Mr. Dunn and TaxACT opposed Intuit’s proposal and believed that it was “probably not legal for that group to restrain trade.” Id. Subsequently, HRB, Intuit and others successfully lobbied the IRS to implement restrictions on the number of taxpayers that could be covered by a free offer through the FFA website. GX 28 (Dunn Dep.) at 114-15; GX 28-4; GX 35 at HRB-DOJ-00912870; GX 569 (DuMars Dep.) at 108, 112-113; Ernst, TT, 9/7/11 a.m., 26-27; GX 41 at 4; GX 25 (TaxHawk Decl.) ¶ 16. HRB desired these restrictions because, among other things, it was concerned about how free-for-all offers would affect the pricing structure for the industry and believed that such offers might undermine the company’s ability to generate money through the paid side of its DDIY business. Ernst, TT, 9/7/11 a.m., at 26-27; GX 531 (Ciaramitaro Dep.) at 60-62; see also 12 GX 41 at 4; GX 25 (TaxHawk Decl.) ¶ 16. The IRS amended the FFA rules in October 2005 to prevent FFA members from making free-for-all offers. Dunn, TT, 9/7/11 p.m., at 78-79; Ernst, TT, 9/7/11 a.m., at 29; GX 42; GX 25 (TaxHawk Decl.) ¶ 16; GX 29 (Intuit Decl.) ¶ 9. Therefore, TaxACT could no longer make its free-for-all offer through the FFA. In tax year 2005, in response to restrictions that the IRS imposed on the scope of offers that could be made through the FFA, TaxACT became the first DDIY company to offer all tax payers a free DDIY product for preparation of federal returns directly on its website. Dunn, TT, 9/7/11 p.m., at 79-80; GX 28 (Dunn Dep.) at 122-23. Today, free offers in various forms are an entrenched part of the DDIY market. Dunn, TT, 9/8/11 a.m., 85; Defs.’ Opening Stmt., TT, 9/6/11 a.m., at 86-87. II. STANDARD OF REVIEW “Section 7 of the Clayton Act, 15 U.S.C. § 18, prohibits a corporation from acquiring ‘the whole or any part of the assets of another [corporation] engaged also in commerce or in any activity affecting commerce, where in any line of commerce or in any activity affecting commerce in any section of the country, the effect of such acquisition may be substantially to lessen competition, or to tend to create a monopoly.” United States v. Sungard Data Sys., Inc., 172 F. Supp. 2d 172, 180 (D.D.C. 2001) (quoting 15 U.S.C. § 18). “The United States is authorized by Section 15 of the Clayton Act to seek an injunction to block a pending acquisition.” Id. (citing 15 U.S.C. § 25). “The United States has the ultimate burden of proving a Section 7 violation by a preponderance of the evidence.” Id. “To establish a Section 7 violation, plaintiff must show that a pending acquisition is reasonably likely to cause anticompetitive effects.” Id. (citing United States v. Penn-Olin Chem. 13 Co., 378 U.S. 158, 171 (1964)); see also United States v. Oracle Corp., 331 F. Supp. 2d 1098, 1109 (N.D. Cal. 2004). “Congress used the words ‘may be substantially to lessen competition’ (emphasis supplied), to indicate that its concern was with probabilities, not certainties.’” FTC v. H.J. Heinz Co., 246 F.3d 708, 713 (D.C. Cir. 2001) (quoting Brown Shoe Co. v. United States, 370 U.S. 294, 323 (1962)). “Section 7 does not require proof that a merger or other acquisition has caused higher prices in the affected market. All that is necessary is that the merger create an appreciable danger of such consequences in the future.” Hosp. Corp. of Am. v. FTC, 807 F.2d 1381, 1389 (7th Cir. 1986). “As this Circuit explained in Heinz, 246 F.3d at 715, the decision in United States v. Baker Hughes Inc., 908 F.2d 981 (D.C. Cir. 1990), sets forth the analytical approach for establishing a Section 7 violation.”5 Sungard, 172 F. Supp. 2d at 180.6 “The basic outline of a section 7 horizontal acquisition case is familiar. By showing that a transaction will lead to undue concentration in the market for a particular product in a particular geographic area, the government establishes a presumption that the transaction will substantially lessen competition.” Baker Hughes, 908 F.2d at 982. To establish this presumption, the government must “show that the merger would produce ‘a firm controlling an undue percentage share of the relevant market, and [would] result [ ] in a significant increase in the concentration of firms in that market.’” Heinz, 246 F.3d at 715 (quoting United States v. Philadelphia Nat’l Bank, 374 U.S. 321, 363 5 Two current Supreme Court justices, in their prior capacities as judges on the Court of Appeals, participated in the D.C. Circuit’s ruling in Baker Hughes. Then-Judge Clarence Thomas wrote the opinion and then-Judge Ruth Bader Ginsburg joined in it. 6 In their closing argument, the defendants chided the government for citing Clayton Act Section 7 cases brought by the Federal Trade Commission for the relevant standard to apply in this case rather than citing to United States v. Baker Hughes Inc., 908 F.2d 981, 982-83 (D.C. Cir. 1990), a case brought by the DOJ. Since this Circuit’s FTC precedents themselves rely heavily on the analytical approach set forth in Baker Hughes, the defendants’ distinction on this point is ultimately of little import. See FTC v. H.J. Heinz Co., 246 F.3d 708, 715 (D.C. Cir. 2001) (“In United States v. Baker Hughes Inc., 908 F.2d 981, 982-83 (D.C.Cir.1990), we explained the analytical approach by which the government establishes a section 7 violation.”). While a lesser showing is required to obtain preliminary relief in an FTC preliminary injunction case, as opposed to a full merits trial like this case, the Court must apply the Baker Hughes analytical framework in either type of Section 7 case. 14 (1963)) (alterations in original). Once the government has established this presumption, the burden shifts to the defendants to rebut the presumption by “show[ing] that the market-share statistics give an inaccurate account of the merger’s probable effects on competition in the relevant market.” Heinz, 246 F.3d at 715 (internal quotation omitted). “‘If the defendant successfully rebuts the presumption [of illegality], the burden of producing additional evidence of anticompetitive effect shifts to the government, and merges with the ultimate burden of persuasion, which remains with the government at all times.’” Id. (quoting Baker Hughes, 908 F.2d at 983). Ultimately, “[t]he Supreme Court has adopted a totality-of-the-circumstances approach to the statute, weighing a variety of factors to determine the effects of particular transactions on competition.” Baker Hughes, 908 F.2d at 984. III. DISCUSSION A. The Relevant Product Market “Merger analysis begins with defining the relevant product market.” FTC v. Swedish Match, 131 F. Supp. 2d 151, 156 (D.D.C. 2000) (citing Brown Shoe, 370 U.S. 294, 324 (1962)). “Defining the relevant market is critical in an antitrust case because the legality of the proposed merger[] in question almost always depends upon the market power of the parties involved.” Id. (quoting FTC v. Cardinal Health, Inc., 12 F. Supp. 2d 34, 45 (D.D.C. 1998)). Indeed, the relevant market definition is often “the key to the ultimate resolution of this type of case because of the relative implications of market power.”7 Id. 7 “A relevant market has two components: (1) the relevant product market and (2) the relevant geographic market. . . . The ‘relevant geographic market’ identifies the geographic area in which the defendants compete in marketing their products or services.” FTC v. CCC Holdings, Inc., 605 F. Supp. 2d 26, 37 (D.D.C. 2009). The parties have stipulated that the relevant geographic market in this case is worldwide. Joint Pre-Hearing Statement ¶ IX, C, 12. DDIY products are provided online and can be used by any individual worldwide – either within the United States or abroad – who needs to prepare and file a U.S. tax return. The products at issue in this case are not used for preparation of foreign tax returns. See Pl.’s Mot. For Prelim. Inj. at 29-30. The Court accepts the parties’ stipulation as to the relevant geographic market. 15 The government argues that the relevant market in this case consists of all DDIY products, but does not include assisted tax preparation or pen-and-paper. Under this view of the market, the acquisition in this case would result in a DDIY market that is dominated by two large players – H&R Block and Intuit – that together control approximately 90 percent of the market share, with the remaining 10 percent of the market divided amongst a plethora of smaller companies. In contrast, the defendants argue for a broader market that includes all tax preparation methods (“all methods”), comprised of DDIY, assisted, and pen-and-paper. Under this view of the market, the market concentration effects of this acquisition would be much smaller and would not lead to a situation in which two firms control 90 percent of the market. This broader view of the market rests primarily on the premise that providers of all methods of tax preparation compete with each other for the patronage of the same pool of customers – U.S. taxpayers. After carefully considering the evidence and arguments presented by all parties, the Court has concluded that the relevant market in this case is, as the DOJ contends, the market for digital do-it-yourself tax preparation products. A “relevant product market” is a term of art in antitrust analysis. The Supreme Court has set forth the general rule for defining a relevant product market: “The outer boundaries of a product market are determined by the reasonable interchangeability of use [by consumers] or the cross-elasticity of demand between the product itself and substitutes for it.” Brown Shoe, 370 U.S. at 325; see also United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377, 395 (1956). In other words, courts look at “whether two products can be used for the same purpose, and, if so, whether and to what extent purchasers are willing to substitute one for the other.” FTC v. Staples, Inc., 970 F. Supp. 1066, 1074 (D.D.C. 1997) (citation omitted); see also Bon-Ton Stores, 16 Inc. v. May Dep’t Stores Co., 881 F. Supp. 860, 868 (W.D.N.Y. 1994) (citing Hayden Pub. Co. v. Cox Broad. Corp., 730 F.2d 64, 71 (2d Cir. 1984)). A broad, overall market may contain smaller markets which themselves “constitute product markets for antitrust purposes.”8 Brown Shoe, 370 U.S. at 325. “[T]he 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.” Staples, 970 F. Supp. at 1075. Traditionally, courts have held that the boundaries of a relevant product market within a broader market “may be determined by examining such practical indicia as industry or public recognition of the [relevant market] as a separate economic entity, the product’s peculiar characteristics and uses, unique production facilities, distinct customers, distinct prices, sensitivity to price changes, and specialized vendors.” FTC v. Whole Foods Market, Inc. 548 F.3d 1028, 1037-38 (D.C. Cir. 2008) (Brown, J.) (quoting Brown Shoe, 370 U.S. at 325).9 See also FTC v. CCC Holdings, Inc., 605 F. Supp. 2d 26, 38 (D.D.C. 2009). These “practical indicia” of market boundaries may be viewed as evidentiary proxies for proof of substitutability and cross-elasticities of supply and demand. Rothery Storage & Van Co. v. Atlas Van Lines, Inc., 792 F.2d 210, 218 (D.C. Cir. 1986). 8 Courts have sometimes referred to such markets-within-markets as “submarkets.” See Brown Shoe, 370 U.S. at 325; Whole Foods, 548 F.3d at 1037-38 (Brown, J.). Other courts and commentators have criticized this “submarket” terminology as unduly confusing, however. See 5C PHILLIP E. AREEDA & HERBERT HOVENKAMP, ANTITRUST LAW ¶ 533, at 251 (3d ed. 2007) (“Courts sometimes describe the closest substitutes as a ‘submarket’ within a larger ‘market’ of less-close substitutes. Although degrees of constraint do in fact vary, the ‘market’ for antitrust purposes is the one relevant to the particular legal issue at hand.”) (internal citations omitted); Geneva Pharms. Tech. Corp. v. Barr Labs, Inc., 386 F.3d 485, 496 (2d Cir. 2004) (“The term ‘submarket’ is somewhat of a misnomer, since the ‘submarket’ analysis simply clarifies whether two products are in fact ‘reasonable’ substitutes and are therefore part of the same market.”). 9 The D.C. Circuit’s decision in Whole Foods lacked a majority opinion. See Whole Foods, 548 F.3d at 1061 n.8 (Kavanaugh, J., dissenting). Judges Brown and Tatel filed separate opinions concurring in the judgment to reverse the District Court and Judge Kavanaugh, in dissent, would have affirmed. See id. at 1032 (Brown, J.); id. at 1041 (Tatel, J.); id. at 1051 n.1 (Kavanaugh, J., dissenting). Thus, in referring to the opinions in Whole Foods, the Court will indicate the name of the Judge whose opinion is cited. 17 An analytical method often used by courts to define a relevant market is to ask hypothetically whether it would be profitable to have a monopoly over a given set of substitutable products. If so, those products may constitute a relevant market. See 5C PHILLIP E. AREEDA & HERBERT HOVENKAMP, ANTITRUST LAW (hereinafter, “Areeda & Hovenkamp”), ¶ 530a, at 226 (3d ed. 2007) (“[A] market can be seen as the array of producers of substitute products that could control price if united in a hypothetical cartel or as a hypothetical monopoly.”). This approach – sometimes called the “hypothetical monopolist test” – is endorsed by the Horizontal Merger Guidelines issued by the DOJ and Federal Trade Commission. See Fed. Trade Comm’n & U.S. Dep’t of Justice Horizontal Merger Guidelines (2010) (hereinafter, “Merger Guidelines”), § 4.1.1.10 In the merger context, this inquiry boils down to whether “a hypothetical profit-maximizing firm, not subject to price regulation, that was the only present and future seller of those products . . . likely would impose at least a small but significant and non-transitory increase in price (“SSNIP”) on at least one product in the market, including at least one product sold by one of the merging firms.” Id. The “small but significant and nontransitory increase in price,” or SSNIP, is typically assumed to be five percent or more. Id. § 4.1.2. Thus, the question here is whether it would be hypothetically useful to have a monopoly over all DDIY tax preparation products because the monopolist could then profitably raise prices for those products by five percent or more; or whether, to the contrary, there would be no reason to monopolize all DDIY tax preparation products because substitution and price competition with other methods of tax preparation would restrain any potential DDIY monopolist from profitably raising prices. In other words, would enough DDIY users switch to the assisted or 10 The Merger Guidelines are not binding upon this Court, but courts in antitrust cases often look to them as persuasive authority. See Staples, 970 F. Supp. at 1081-82. 18 pen-and-paper methods of tax preparation in response to a five-to-ten percent increase in DDIY prices to make such a price increase unprofitable? In evaluating the relevant product market here, the Court considers business documents from the defendants and others, the testimony of the fact witnesses, and the analyses of the parties’ expert economists. This evidence demonstrates that DDIY is the relevant product market in this case. 1. The Defendants’ Documents Show That DDIY Is The Relevant Product Market. When determining the relevant product market, courts often pay close attention to the defendants’ ordinary course of business documents. See, e.g., Staples, 970 F. Supp. at 1076; CCC Holdings, 605 F. Supp. 2d at 41-42. The government argues that the defendants’ ordinary course of business documents in this case “conclusively demonstrate that competition with other [DDIY] firms drive Defendants’ pricing decisions, quality improvements, and corporate strategy” for their own DDIY products—thus supporting the government’s view of the relevant market. Pl.’s Post-Trial Mem. at 7. The defendants contend that the government has relied on “select, ‘out-of-context’ snippets from documents,” and that the documents as a whole support the defendants’ view that the relevant product market is all methods of tax preparation. Defs.’ Post-Trial Mem. at 1. The Court finds that the documentary evidence in this case supports the conclusion that DDIY is the relevant product market. Internal TaxACT documents establish that TaxACT has viewed DDIY offerings by HRB and TurboTax as its primary competitors, that it has tracked their marketing, product offerings, and pricing, and that it has determined its own pricing and business strategy in relation to those companies’ DDIY products. See GX 295-16 (“Competitive Analysis” comparing the three companies); GX 102 (email explaining TaxACT is a “direct competitor” with HRB and Intuit’s 19 products); GX 55 (Greif Dep.) at 137-38 (describing TaxACT’s compilation of a routine, end-ofseason competitive analysis that “typically” covers Intuit, HRB, and TaxACT). Confidential memoranda prepared by TaxACT’s investment bankers for potential private equity buyers of TaxACT identify HRB and TurboTax as TaxACT’s primary competitors in a DDIY market. See GX 7 (Greene Holcomb & Fisher “Confidential Memorandum”) at 14 (“The Company’s major competitors for both desktop and Internet-based income tax software and e-filing services include Intuit (the makers of TurboTax software) and H&R Block (the makers of TaxCut software).”); GX 134 (Deutsche Bank “Confidential Information Memorandum”) at 17 (“The Company’s two main competitors, Intuit and H&R Block. . .”); see also Dunn, TT, 9/7/11 p.m., at 97-104. These documents also recognize that TaxACT’s strategy for competing with Intuit and HRB is to offer a lower price for what it deems a superior product. GX 7 at 14 (“Relative to its two major competitors, 2nd Story has positioned its product offerings as being of equal or higher quality, and completely fulfilling the needs of a vast portion of the potential market. It also pursues a pricing strategy that positions its products and services meaningfully below either Intuit or H&R Block, in some instances free.”). While, as defendants point out, parts of these TaxACT documents also discuss the broader tax preparation industry, these documents make clear that TaxACT’s own view – and that conveyed by its investment bankers to potential buyers – is that the company primarily competes in a DDIY market against Intuit and HRB and that it develops its pricing and business strategy with that market and those competitors in mind. These documents are strong evidence that DDIY is the relevant product market. See Whole Foods, 548 F.3d at 1045 (Tatel, J.) (“[E]vidence of industry or public recognition of the submarket as a separate economic unit 20 matters because we assume that economic actors usually have accurate perceptions of economic realities.”) (internal quotation omitted). Internal HRB documents also evidence HRB’s perception of a discrete DDIY market or market segment. HRB and its outside consultants have tracked its digital competitors’ activities, prices, and product offerings. See GX 28-19 (“2009 Competitive Price Comparison”); GX 118 (independent analyst’s report analyzing digital competitors as one of three separate categories of competitors); GX 61-8 at 1 (slide on competition in “digital market” identifying TurboTax and TaxACT as competitors); GX 199 (HRB “digital strategy update” Powerpoint tracking features and prices for TurboTax and TaxACT); GX 188 (HRB spreadsheet comparing HRB, TurboTax, and TaxACT prices for various product offerings). Documents from HRB’s DDIY business have also referred to HRB, TaxACT, and TurboTax as the “Big Three” competitors in the DDIY market. GX 61-3 (“OCS Offsite Competitive Intelligence Review of TS07”) at 5; GX 61-4 at 1 (email referencing request for data from consultant regarding “big 3 digital tax prep companies”); see also GX 70 (email from head of HRB’s digital business stating its “only real direct competitors are turbotax in san diego and taxact in cedar rapids” [sic]); Ernst, TT, 9/7/11 a.m., at 13-14. Finally, the documents show that, in connection with a proposed acquisition of TaxACT, HRB identified the proposed transaction as a way to grow its digital “market share” and has measured TaxACT’s market share in a DDIY market. GX 130 at 96-99; GX 21-37 (projections from 2009 for different potential scenarios for acquisition of TaxACT, including their effect on DDIY market share); see also Newkirk, TT, 9/7/11 a.m., at 95-96 (explaining GX 21-37). All of these documents also provide evidence that DDIY is a relevant product market. The defendants acknowledge that “the merging parties certainly have documents that discuss each other and digital competitors generally, and even reference a digital market and the 21 ‘Big Three,’” but contend this evidence is insufficient to prove a market. Defs.’ Post-Trial Mem. at 9. Rather, the defendants argue that the documents show that the relevant market is all methods of tax preparation, especially in light of documented competition between DDIY providers and assisted providers for the same overall pool of U.S. taxpayers who are potential customers. See id. 9-10; see, e.g., DX 78 at 4 (Intuit document explaining 2011 strategic goal of acquiring tax store customers); GX 650 at 41 (Intuit document noting goal of acquiring tax store customers and specifically mentioning HRB). As discussed below, the Court disagrees and finds that the relevant product market is DDIY products. 2. The Relevant Product Market Does Not Include Assisted Tax Preparation Or Manual Preparation. It is beyond debate – and conceded by the plaintiff – that all methods of tax preparation are, to some degree, in competition. Pl.’s Post-Trial Mem. at 8. All tax preparation methods provide taxpayers with a means to perform the task of completing a tax return, but each method is starkly different. Thus, while providers of all tax preparation methods may compete at some level, this “does not necessarily require that [they] be included in the relevant product market for antitrust purposes.” Staples, 970 F. Supp. at 1075. DDIY tax preparation products differ from manual tax preparation and assisted tax preparation products in a number of meaningful ways. As compared to manual and assisted methods, DDIY products involve different technology, price, convenience level, time investment, mental effort and type of interaction by the consumer. Taken together, these different attributes make the consumer experience of using DDIY products quite distinct from other methods of tax preparation. See Whole Foods, 548 F.3d at 1037-38 (Brown, J.) (noting that a “product’s peculiar characteristics and uses” and “distinct prices” may distinguish a relevant market) (citing Brown Shoe, 370 U.S. at 325); see also, e.g., GX 130 at 140 (HRB internal analysis discussing convenience and price as factors differentiating DDIY and 22 assisted methods for consumers). The question for this court is whether DDIY and other methods of tax preparation are “reasonably interchangeable” so that it would not be profitable to have a monopoly over only DDIY products. a. Assisted Tax Preparation Is Not In The Relevant Product Market. Apart from the analysis of their economic expert, the defendants’ main argument for inclusion of assisted tax preparation in the relevant market is that DDIY and assisted companies compete for customers.11 As evidence for this point, the defendants emphasize that Intuit’s marketing efforts have targeted HRB’s assisted customers. See DX 78 at 3 (Intuit document noting strategic goal to “Beat Tax Store[s]”). While the evidence does show that companies in the DDIY and assisted markets all generally compete with each other for the same overall pool of potential customers – U.S. taxpayers – that fact does not necessarily mean that DDIY and assisted must be viewed as part of the same relevant product market. DDIY provides customers with tax preparation services through an entirely different method, technology, and user experience than assisted preparation. As Judge Tatel explained in Whole Foods: [W]hen the automobile was first invented, competing auto manufacturers obviously took customers primarily from companies selling horses and buggies, not from other auto manufacturers, but that hardly shows that cars and horse-drawn carriages should be treated as the same product market. That Whole Foods and Wild Oats have attracted many customers away from conventional grocery stores by offering extensive selections of natural and organic products thus tells us nothing about whether Whole Foods and Wild Oats should be treated as operating in the same market as conventional grocery stores. Indeed, courts have often found that sufficiently innovative retailers can constitute a distinct product market even when they take customers from existing retailers. Whole Foods, 548 F.3d at 1048; see also Staples, 970 F. Supp. at 1074-80 (finding a distinct market of office supply superstores despite competition from mail-order catalogues and stores carrying a broader range of merchandise). 11 The defendants’ primary argument for inclusion of both assisted and pen-and-paper in the relevant market is based upon their economic expert’s analysis of data derived from two consumer surveys commissioned by the defendants. The Court will analyze the arguments of the defendants’ expert economist separately below. 23 The key question for the Court is whether DDIY and assisted products are sufficiently close substitutes to constrain any anticompetitive DDIY pricing after the proposed merger. Evidence of the absence of close price competition between DDIY and assisted products makes clear that the answer to that question is no—and that DDIY is the relevant product market here. See Swedish Match, 131 F. Supp. 2d at 165 (“Distinct pricing is also a consideration” in determining the relevant product market) (citing Brown Shoe, 370 U.S. at 325). Significantly, despite some DDIY efforts to capture tax store customers, none of the major DDIY competitors sets their prices based on consideration of assisted prices. See, e.g., Ernst, TT, 9/7/11 a.m., at 35 (HRB set its digital and assisted prices separately); {redacted} (Dep.) at 183:18-25 (explaining that {redacted} does not consider assisted pricing in setting prices because its prices are already “substantially less than both tax stores and most professionals”). Indeed, there are quite significant price disparities between the average prices of DDIY and assisted products. The average price of TurboTax, the most popular DDIY brand is approximately $55. GX 293 (Intuit Dep.) at 21. The average price of HRB’s DDIY products is approximately $25. GX 296-7 at 6. Overall, the DDIY industry average price is $44.13. GX 121 at 57. In contrast, the typical price of an assisted tax return is significantly higher, in the range of $150-200.12 A 10 percent or even 20 percent price increase in the average price of DDIY would only move the average price up to $48.54 or $52.96, respectively – still substantially below the average price of assisted tax products. The overall lack of evidence of price competition between DDIY and assisted products supports the conclusion that DDIY is a separate relevant product market for evaluating this transaction, despite the fact that DDIY and assisted firms target their marketing efforts at the same pool of customers. 12 See GX 128 (HRB “TS10 Market Dynamics” presentation) at 38 {redacted}; see also id. {redacted}; GX 293 (Intuit Dep.) at 21:9-14 (“The average price of a tax store is in the range of $200.”); Bennett, TT, 9/6/11 p.m., at 100 (estimating $150 range for assisted returns offered at Jackson Hewitt and HRB offices at Wal-Mart locations). 24 The defendants point to some evidence that HRB sets prices for certain assisted products to compete with DDIY. For example, defendants note that in 2009, HRB “reduced prices on its assisted tax preparation services to $39 for federal 1040EZ preparation and $29 for state tax preparation to compete with and {redacted}” to DDIY. DFF ¶ 77a. These are limited product offerings for which prices appear well below even the 25th percentile price for HRB’s assisted products. See GX 128 (HRB “TS10 Market Dynamics” presentation) at 38 (noting, for Tax Season 2010, that the 25th percentile for prices at HRB stores was {higher than DDIY}). Relatedly, the defendants’ claim that prices for assisted and DDIY products “significantly overlap” is not strongly supported and relies on a comparison of the most limited, low-end assisted products with DDIY products generally. See DFF ¶ 78b (citing tax year 2009 data that show that 14 percent of customers using name-brand tax stores paid $50 or less and another 20 percent paid between $51-100); id. ¶ 78c-d (quoting prices for Jackson Hewitt’s preparation of form 1040EZ, a simplified tax form, at Wal-Mart and for HRB’s Second Look service, which actually only double-checks an already completed tax return for errors). In sum, while defendants’ have identified isolated instances in which assisted product offerings are priced lower than the average prices for typical assisted products, they do not and cannot demonstrate that this is generally the case. Testimony from HRB executives further supports treating DDIY as a relevant product market in evaluating this transaction. HRB’s DDIY and assisted businesses are run as separate business units. Bennett, TT, 9/6/11 a.m., at 106. Alan Bennett, who was the CEO of HRB in 2010 when the parties reached the proposed merger agreement, testified that “net-net,” he did not believe that HRB’s DDIY business had impacted its assisted business in terms of taking away 25 customers.13 Id. at 108; see also GX 1151 at 4 (HRB internal analysis stating “Online is not growing materially at the expense of assisted.”). Mark Ernst, HRB’s CEO from 2001 to 2007, also explained that, in his opinion based on research he reviewed while at HRB, the primary reason consumers switched between assisted and DDIY was because of “life events” that led to changes in tax status. Ernst, TT, 9/7/11 a.m., at 34-35. Finally, defendants argue that their broad relevant market is appropriate because there is “industry movement toward ‘hybrid’ products that combine some elements of both digital and assisted tax preparation.” Defs.’ Post-Trial Mem. at 11. Based on the evidence presented at the hearing, however, it would be premature for the Court to identify any trend toward hybrid products. In fact, neither Intuit nor TaxACT presently offers a hybrid product and the defendants openly concede that HRB’s current hybrid product has had “somewhat limited success,” which defendants attribute to “technical issues” and a “lack of consistent marketing.” Id. at 11 n.16. {redacted} {T}he Court finds it unlikely that there will be a sufficiently large scale shift into these products in the immediate future to compel the conclusion that DDIY and assisted products make up the same relevant product market. b. Manual Tax Preparation Is Not In The Relevant Product Market. The defendants also argue that manual tax preparation, or pen-and-paper, should be included in the relevant product market. At the outset, the Court notes that pen-and-paper is not a “product” at all; it is the task of filling out a tax return by oneself without any interactive assistance. Even so, the defendants argue pen-and-paper should be included in the relevant product market because it acts as a “significant competitive constraint” on DDIY. Defs.’ Post- 13 By “net-net,” Mr. Bennett meant that while there is customer switching between the DDIY and assisted businesses, the total share of customers in each has been relatively stable over the past few years, such that Mr. Bennett could conclude that the two business lines “do not steal customers back and forth net.” Bennett, TT, 9/6/11 a.m., at 108. 26 Trial Mem. at 11. The defendants’ argument relies primarily on two factors. First, the defendants’ cite the results of a 2011 email survey of TaxACT customers. See id. For reasons detailed in the following section, the Court declines to rely on this email survey. Second, the defendants point to documents and testimony indicating that TaxACT has considered possible diversion to pen-and-paper in setting its prices. See id. at 11-12. The Court finds that pen-and-paper is not part of the relevant market because it does not believe a sufficient number of consumers would switch to pen-and-paper in response to a small, but significant increase in DDIY prices. The possibility of preparing one’s own tax return necessarily constrains the prices of other methods of preparation at some level. For example, if the price of DDIY and assisted products were raised to $1 million per tax return, surely all but the most well-heeled taxpayers would switch to pen-and-paper. Yet, at the more practical price increase levels that trigger antitrust concern – the typical five to ten percent price increase of the SSNIP test – pen-and-paper preparation is unlikely to provide a meaningful restraint for DDIY products, which currently sell for an average price of $44.13. GX 121 at 57. The government well illustrated the overly broad nature of defendants’ proposed relevant market by posing to the defendants’ expert the hypothetical question of whether “sitting at home and drinking chicken soup [would be] part of the market for [manufactured] cold remedies?” Meyer, TT, 9/13/11 a.m., at 65. The defendants’ expert responded that the real “question is if the price of cold medicines went up sufficiently, would people turn to chicken soup?” Id. As an initial matter, in contrast to the defendants’ expert, the Court doubts that it would ever be legally appropriate to define a relevant product market that included manufactured cold remedies and ordinary chicken soup. This conclusion flows from the deep functional differences between those products. Setting that issue aside, however, a price has increased “sufficiently” to trigger 27 antitrust concern at the level of a five to ten percent small, but significant non-transitory increase in price. Just as chicken soup is unlikely to constrain the price of manufactured cold remedies sufficiently, the Court concludes that a SSNIP in DDIY would not be constrained by people turning to pen-and-paper. First, the share of returns prepared via pen-and-paper has dwindled over the past decade, as the DDIY market has grown. Bennett, TT, 9/6/11 a.m, at 118; GX 296 (Houseworth Dep.) at 66-68. Second, while pen-and-paper filers have been a net source of new customers for DDIY companies, both HRB and {redacted} executives have testified that they do not believe their DDIY products compete closely with pen-and-paper methods. {redacted} (Dep.) at 37:20-38:10; see GX 296 (Houseworth Dep.) 89-90. Third, courts in antitrust cases frequently exclude similar “self-supply” substitutes from relevant product markets. See, e.g., FTC v. H.J. Heinz Co., 116 F. Supp. 2d 190, 195 (D.D.C. 2000), rev’d on other grounds, 246 F.3d 708 (D.C. Cir. 2001) (noting that homemade baby food and breast milk should not be included in the jarred baby food market even though substitution was possible because “the Supreme Court’s interchangeability test refers to products.”); CCC Holdings, 605 F. Supp. 2d at 41-42 (excluding books that can be used to perform insurance loss valuations by hand from market for loss valuation software); United States v. Visa U.S.A. Inc., 163 F. Supp. 2d 322, 338 (S.D.N.Y. 2001) (excluding cash and checks from general purpose credit card market). The main case the defendants rely on to show that “self-supply” substitutes should be included in the relevant market involved a consumer market consisting of vertically integrated companies and explicitly distinguished cases, such as this one, involving markets of individual consumers. In United States v. Sungard Data Systems, Inc., Judge Huvelle found that disaster recovery computer systems developed internally by companies were in the same relevant product market as shared data recovery systems provided by outside vendors. Sungard, 172 F. Supp. 2d 28 at 187-89. The Sungard court, however, distinguished the case before it – which involved vertical integration – from the situation in Heinz, the case involving the market for jarred baby food, because “homemade baby food is not an aspect of vertical integration . . . [and] individual consumers cannot vertically integrate by producing a product that they would otherwise have to purchase.” Id. at 187 n.15. In finding that in-house computer systems were included in its relevant product market, the Sungard court cited the following example from Areeda & Hovenkamp ¶ 535e regarding vertical integration: If iron ore is the relevant market and if shares are best measured there by sales, then internally used ore—so-called captive output—is part of the ore market even though it is not sold as such. In measuring the market power of a defendant selling iron ore, the ore used internally by other firms constrains the defendant’s ability to profit by raising ore prices to monopoly levels. The higher ore price may induce an integrated firm to expand its ore production—to supply others in direct competition with the alleged monopolist or to expand its own steel production and thereby reduce the demand of other steel makers for ore, or both. Hence, captive output constrains the defendant regardless of whether integrated firms sell their ore to other steel makers previously purchasing from the defendant. In sum, the integrated firm’s ore output belongs in the market. Id. at 186 n.14. This rationale for including “self-supply” in a relevant product market does not appear to apply to the DDIY market in which the consumers are individuals and not also potential traders or producers. While some diversion from DDIY to manual filing may occur in response to a SSNIP, the Court finds that it would likely be limited and marginal. The functional experience of using a DDIY product is meaningfully different from the self-service task of filling out tax forms independently. Manual completion of a tax return requires different tools, effort, resources, and time investment by a consumer than use of either DDIY or assisted methods. The following discussion from United States v. Visa U.S.A. Inc. regarding why cash and checks should not be included in the credit card market is instructive here: 29 [A]lthough it is literally true that, in a general sense, cash and checks compete with general purpose cards as an option for payment by consumers and that growth in payments via cards takes share from cash and checks in some instances, cash and checks do not drive many of the means of competition in the general purpose card market. In this respect, [the expert’s] analogy of the general purpose card market to that for airplane travel is illustrative. [The expert] argues that while it is true that at the margin there is some competition for customers among planes, trains, cars and buses, the reality is that airplane travel is a distinct product in which airlines are the principal drivers of competition. Any airline that had monopoly power over airline travel could raise prices or limit output without significant concern about competition from other forms of transportation. The same holds true for competition among general purpose credit and charge cards. Visa U.S.A. Inc., 163 F. Supp. 2d at 338. Here, the same analogy to airplane travel holds true for competition among DDIY providers, who provide a distinct product for completion of tax returns. Indeed, the pen-and-paper method, in which the consumer essentially relies on his or her own labor to prepare a tax return, is perhaps most analogous to walking as opposed to purchasing a ride on any means of transportation. In sum, filling out a tax return manually is not reasonably interchangeable with DDIY products that effectively fill out the tax return with data input provided by the consumer. Inclusion of all possible methods of tax preparation, including pen-and-paper, in the relevant product market also violates the principle that the relevant product market should ordinarily be defined as the smallest product market that will satisfy the hypothetical monopolist test. See Merger Guidelines § 4.1.1 (“When the Agencies rely on market shares and concentration, they usually do so in the smallest relevant market satisfying the hypothetical monopolist test.”); see also Warren-Boulton, TT, 9/8/11 p.m., at 35-36. Indeed, the defendants’ inclusion of pen-and-paper in the relevant market ignores at least one obvious, smaller market possibility that they might have proposed – the combined market of all DDIY and assisted tax preparation products. It is hardly plausible that a monopolist of this market – to which the only alternative would be pen-and-paper – could not impose a SSNIP. 30 The defendants’ proposed relevant market of all methods of tax return preparation is so broadly defined that, as the plaintiff’s expert testified, there are no conceivable alternatives besides going to jail, fleeing to Canada, or not earning any taxable income. Warren-Boulton, TT, 9/8/11 p.m., at 35-36. As the plaintiff’s expert put it, “if you’re talking about the market for all tax preparation, you’re talking about a market where, in economist terms, demand is completely [in]elastic. There are no alternatives.” Id. at 35. In such circumstances, the usual tools of antitrust analysis – such as the hypothetical monopolist test – cease being useful because it is self-evident that a monopolist of all forms of tax preparation, including self-preparation, could impose a small, but significant price increase. Indeed, a monopolist in that situation could essentially name any price since taxpayers would have no alternative but to pay it. As the plaintiff’s expert testified, defining a market that broadly negates the entire purpose of defining a relevant market in an antitrust case. You want to define a relevant market in an antitrust case so then [you can calculate] shares and the change in shares makes sense. I don’t want to go to infinity . . . I want to define a relevant market under . . .the smallest market principle, which is I want to define the relevant market so that if a hypothetical monopolist . . . did manage to control all of those products, they would impose a significant price increase, large enough to be of concern but not so large as to make the whole exercise pointless. Id. at 35-36. The Court agrees with this assessment and finds the defendants’ proposed relevant market to be overbroad. 3. The Economic Expert Testimony Tends To Confirm That DDIY Is The Relevant Product Market. Both the plaintiff and the defendants presented testimony from expert economists to support their view of the relevant product market.14 In addition to their testimony at the hearing, 14 The plaintiff presented expert testimony on market definition from Frederick R. Warren-Boulton, an economist at MiCRA, an economics consulting and research firm. GX 121 (Warren-Boulton Rep.) at 1. Dr. Warren-Boulton holds a B.A. from Yale University, a Ph.D. in economics from Princeton University, and formerly served as the chief economist for the Antitrust Division of the U.S. Department of Justice. Id. Dr. Warren-Boulton has previously served as an expert witness in other antitrust cases, including cases challenging the possible 31 these expert witnesses also provided a detailed expert report and an affidavit summarizing their analysis and conclusions. The Court finds that the analysis performed by the plaintiff’s expert tends to confirm that DDIY is a relevant product market, although the available data in this case limited the predictive power of the plaintiff’s expert’s economic models. The Court also finds that it cannot draw any conclusions from defendants’ expert’s analysis because of severe shortcomings in the underlying consumer survey data upon which the defendants’ expert relied. a. Plaintiff’s Expert - Dr. Warren-Boulton The plaintiff’s expert, Dr. Warren-Boulton, found the relevant product market to be DDIY. He determined that a hypothetical monopolist of DDIY products could profitably impose a SSNIP for at least one DDIY product, and that consumer substitution to assisted methods or pen-and-paper would be insufficient to defeat the SSNIP. GX 121 (Warren-Boulton Rep.) at 12. Dr. Warren-Boulton began his analysis by postulating that DDIY was the relevant product market and then he used two principal analytical tests to confirm the validity of that assumption. He began by testing DDIY as a relevant market for a few reasons. First, he concluded that the parties’ DDIY products are substantially similar in terms of functionality. GX 121 (Warren-Boulton Rep.) at 12-18. Second, he concluded from his review of the defendants’ business documents that they viewed DDIY as a discrete product market when competing in the anticompetitive effects of a merger or acquisition. Id. (noting involvement in FTC v. Staples, Inc., 970 F. Supp. 1066, 1074 (D.D.C. 1997)). The defendants presented expert testimony from Christine Siegwarth Meyer, an economist at National Economic Research Associates, Inc., an economics consulting and research firm. DX 17 (Meyer Rep.) at 1. Dr. Meyer holds a B.A. from the United States Military Academy at West Point, a Ph.D. in economics from the Massachusetts Institute of Technology, and has taught economics at the university level. Id. Dr. Meyer has not previously provided expert testimony regarding the possible anticompetitive effects of a merger or acquisition. Meyer, TT, 9/13/11 a.m., at 39. 32 ordinary course of business. Id. Third, he ruled out including pen-and-paper and assisted products in the relevant product market based on a consideration of various data. Id. at 24-32. Dr. Warren-Boulton’s decision to begin the relevant market analysis with DDIY was appropriate. See Areeda & Hovenkamp ¶ 536, at 287 (“[T]wo products are provisionally part of the same market [for hypothetical monopolist analysis] when they employ similar technologies and similar costs and customers use them interchangeably. . . In cases of doubt, [products] should generally be excluded from the provisional market, for incorrect exclusions will ultimately be brought into the market via the price increase methodology.”). The parties’ DDIY products all provide a fundamentally similar service and a similar user experience for the consumer when compared with other methods of tax preparation. The DDIY consumer sits down at a computer and interacts with the DDIY software, which prompts the consumer for information and ultimately completes the consumer’s tax return. This experience is qualitatively different than that of hiring a tax professional or figuring out how to complete one’s own tax return manually. Various other evidence in the record also supports the fundamental functional similarity of the technology underlying the parties’ DDIY products – perhaps most notably the testimony that post-merger, HRB plans to migrate {redacted} onto TaxACT’s software “engine” {redacted}. See Dunn, TT, 9/8/11 p.m. (sealed), at 16-17. As discussed in detail above, various documentary evidence suggests that the parties treat DDIY as a distinct product market in the ordinary course of business. Dr. Warren-Boulton also considered whether the pen-and-paper and assisted methods should be included in the provisional relevant market, as the defendants contend, and concluded that they should not be. 33 Dr. Warren-Boulton ruled out including pen-and-paper in the relevant product market, concluding instead that historical tax return data reflects “a gradual migration of customers to [DDIY] from more traditional methods like pen-and paper.” GX 121 (Warren-Boulton Rep.) at 24. The percentage of returns prepared by pen-and-paper has fallen considerably over the last decade, while the percentage of DDIY has grown. Id. Changes in the yearly percentage shares of taxpayers using pen-and-paper do not appear correlated to changes in the yearly average price of DDIY. Id. at 27. Finally, based on IRS data, Dr. Warren-Boulton observed that taxpayers who switched from DDIY to pen-and-paper for tax seasons 2008 and 2009 on average experienced a decrease in tax return complexity, suggesting that much switching from DDIY products to pen-and-paper is driven by such complexity decreases.15 Dr. Warren-Boulton also ruled out including the assisted tax preparation methods in the relevant market based on consideration of several factors. He reviewed HRB documents that conclude that growth in DDIY has not come at the expense of HRB’s assisted business. Id. at 28. Testimony from HRB employees, including the former CEO, also reinforced the same conclusion. Id. at 28-29. He also cited HRB internal studies, which concluded that consumers who have switched from DDIY to assisted are likely to have experienced a change in tax complexity. He found that HRB’s internal conclusion was consistent with IRS switching data, which also indicated a correlation between switching from DDIY to assisted and an increase in tax complexity. Id. at 29-30. Finally, Dr. Warren-Boulton noted that, based on data from tax years 2004-2009, increases in the relative price of assisted products were not associated with decreases in the relative market share of assisted products and increases in the relative market 15 Switching, as discussed further below, refers to the switching of consumers between different products for any reason. The IRS categorizes tax returns into one of three complexity categories: Simple, Intermediate, and Complex. Accordingly, the IRS data only reflects complexity changes that are sufficient to result in assignment to a different one of the three categories. 34 share of DDIY, as might be expected if DDIY and assisted prices moved in a single, priceresponsive market. Id. at 32. Therefore, having determined that the best provisional relevant market is DDIY and not all methods of tax preparation, Dr. Warren-Boulton then performed two economic tests to confirm that a hypothetical monopolist of all DDIY products could profitably impose a SSNIP. If these economic tests indicated that a hypothetical monopolist could not profitably impose a SSNIP, then the tests would call for the relevant market to be expanded. The tests, however, validated the relevant market as DDIY, as detailed below. The economic tests Dr. Warren-Boulton applied relied heavily upon switching data from the IRS. Switching refers to the number of consumers who switch between different products for any reason. In any given year, many taxpayers switch from the tax preparation method they used in the prior year to a new method. Since the IRS processes all U.S. tax returns each year and tracks data about the methods of tax preparation that taxpayers used, there is ample, reliable data that market analysts can use to see how many taxpayers switched between methods each year. The IRS data, however, provides little direct insight about why any given taxpayer switched methods of preparation. The switch could have been for reasons of price, convenience, changes in the consumer’s personal situation, an increase or decrease in tax complexity, a loss of confidence in the prior method of preparation, or any other reason. As opposed to switching, diversion refers to a consumer’s response to a measured increase in the price of a product. In other words, diversion measures to what extent consumers of a given product will switch (or be “diverted”) to other products in response to a price increase in the given product. The IRS switching data does not directly measure diversion because switching can occur for any number of reasons, many of which may not involve price. 35 Unfortunately, no direct, reliable data on diversion exists in this case. The plaintiff’s expert argues, however, that the IRS switching data can provide at least some estimate of diversion. While this approach is not without its limitations, as discussed further below, the Court finds that the switching data is at least somewhat indicative of likely diversion ratios. Moreover, the IRS data is highly reliable because (1) the sample size is enormous, since it encompasses over 100 million taxpayers, and (2) the data reflects actual historical tax return filing patterns as opposed to predicted behavior.16 The defendant’s expert, who criticizes reliance on this switching data, suggests instead that a better analysis can be based upon simulated diversion data derived from consumer surveys commissioned by the defendants. As described more fully below, however, the shortcomings of these survey-derived diversion data are so substantial that the Court cannot rely on them. i. Critical Loss Analysis The first economic test Dr. Warren-Boulton performed is known as a “critical loss” analysis. This test attempts to calculate “the largest amount of sales that a monopolist can lose before a price increase becomes unprofitable.” Swedish Match, 131 F. Supp. 2d at 160. Dr. Warren-Boulton calculated that for a 10 percent price increase in DDIY, the price increase would 16 One limitation in the IRS data set is that if a taxpayer uses a DDIY product to prepare the return, but then prints and mails the return instead of e-filing it, the IRS does not attribute the filing to the DDIY provider and instead lists it in a generic “v-coded” pool of returns. At the hearing, the defendants’ criticized the IRS switching data set as problematic on these grounds, suggesting that up to 30 million returns may be “v-coded.” See Warren-Boulton, TT 9/20/11 a.m., at 21-22. As Dr. Warren-Boulton fully addressed in his expert report, however, a “conservative method for dealing with this issue is to drop all v-coded returns from the analysis,” which would still leave well over 100 million returns in the IRS data set. Id.; GX 121 (Warren-Boulton Rep.) at 47. The defendants did not identify any reason the v-coded data would be likely to skew the data set. Thus, even if the v-coded data is disregarded, the IRS data set remains extensive and reliable. It is also worth noting that the IRS data does not distinguish between the DDIY providers’ various products, so only firm-level switching rates are available. GX 121 (Warren-Boulton Rep.) at 47 36 be profitable if the resulting lost sales did not surpass 16.7 percent.17 GX 121 (Warren-Boulton Rep.) at 34. Dr. Warren-Boulton then sought to compare this critical loss threshold with “aggregate diversion ratios.” The aggregate diversion ratio for any given product represents the proportion of lost sales that are recaptured by all other firms in the proposed market as the result of a price increase. Since these lost sales are recaptured within the proposed market, they are not lost to the hypothetical monopolist. According to Dr. Warren-Boulton, economists have shown that if the aggregate diversion ratio to products inside the proposed relevant market exceeds the critical loss threshold, then the critical loss analysis indicates that a SSNIP at that level would be profitable for a hypothetical monopolist. Id. at 34 (citing Michael Katz and Carl Shapiro, Critical Loss: Let’s Tell the Whole Story, ANTITRUST (Spring 2003) at 49 -56); see also WarrenBoulton, TT, 9/9/11 p.m., at 33-34. Because no diversion data is available, Dr. Warren-Boulton relied instead on IRS switching data to estimate aggregate diversion ratios. Id. These data show that of the taxpayers who left HRB’s DDIY products between tax year 2007 and 2008,18 57 percent went to other DDIY providers. Of those who left TaxACT, 53 percent stayed in DDIY, and for TurboTax, 39 percent stayed in DDIY. Id. at 34-35. Since these numbers are all well above the 16.7 percent critical loss threshold, Dr. Warren-Boulton concluded a 10 percent SSNIP in the DDIY market would be profitable for a hypothetical monopolist. In cross-examining Dr. Warren-Boulton, the defendants suggested that the critical loss test is meaningless because it would seem to validate numerous different candidate markets 17 The formula for critical loss is L = X/(X + M), where L is the critical loss, X is the percentage price increase, and M is the hypothetical monopolist’s gross margin. Assuming a 50 percent margin, which Dr. Warren-Boulton claims is a conservative estimate for firms in the DDIY market, then the critical loss for a 10 percent SSNIP is 16.7 percent. 16.7 percent is the result of applying 10 percent and 50 percent in the formula X/(X+M): .167 = .1/(.1+.5). 18 These are the last two years for which this data was available. 37 consisting of various assortments of tax preparation businesses. Warren-Boulton, TT, 9/9/11 p.m., at 20-42. For example, the defendants demonstrated that the test could also validate a market consisting of just HRB and Intuit or a market consisting of just TaxACT and Intuit. See DX 9802. Dr. Warren-Boulton noted in his testimony, however, that such markets are “smaller, irrelevant” markets for evaluating the proposed transaction between HRB and TaxACT. Warren-Boulton, TT, 9/9/11 p.m., at 41; see also Areeda & Hovenkamp ¶ 533c, at 254 (“[C]ourts correctly search for a ‘relevant market’ – that is a market relevant to the particular legal issue being litigated.”). The fact that critical loss analysis would validate other groupings of businesses does not undermine Dr. Warren-Boulton’s reliance on it to validate DDIY as the relevant market in this case.19 Indeed, rather than urging a smaller relevant market definition, the defendants urged the Court to define the market much more broadly. Nonetheless, the Court appreciates the defendants’ point that the critical loss test alone cannot answer the relevant market inquiry. While some inappropriate proposed relevant markets would be ruled out by the critical loss test, the fact that the test could still confirm multiple relevant markets means that the Court must rely on additional evidence in reaching the single, appropriate market definition. ii. Merger Simulation In addition to the critical loss analysis, Dr. Warren-Boulton also performed an economic simulation of a merger among the HRB, TaxACT, and Intuit. GX 121 (Warren-Boulton Rep.) at 19 The defendants also referred obliquely in cross examination to an academic debate surrounding the proper way to perform critical loss analysis. Warren-Boulton, TT, 9/9/11 p.m., at 23. Dr. Warren-Boulton acknowledged his awareness of the existence of this debate and the defendants’ counsel did not pursue the topic further. Id. The Court has no basis for disputing Dr. Warren-Boulton’s application of critical loss analysis based merely on the existence of unspecified academic critiques. The Court notes that the critical loss analysis is specifically endorsed by the Merger Guidelines as a method for implementing the SSNIP test, see Merger Guidelines § 4.1.3, and has been accepted by courts as a standard methodology. See FTC v. CCC Holdings Inc., 605 F. Supp. 2d 26, 40 n.16 (D.D.C. 2009) (“Critical loss analysis is a standard tool used by economists to study potentially relevant markets.”). The court in CCC Holdings ultimately did not rely on the expert’s application of critical loss analysis due to what the court deemed a “gap” or oversight in the expert’s reasoning, but the court nonetheless adopted the same relevant product market that the critical loss analysis had validated. See id. at 40-41. 38 35. This simulation, known as a Bertrand model, predicted that a monopolist of the DDIY products of these three companies would find it profit-maximizing to raise TaxACT’s price by 83 percent, HRB’s price by 37 percent and TurboTax’s price by 11 percent absent efficiencies. Id. Dr. Warren Boulton concluded that this simulation also confirms that DDIY is the relevant product market.20 iii. Critiques of Dr. Warren-Boulton’s Analysis The defendants’ expert, Dr. Meyer, critiques Dr. Warren-Boulton’s analysis in numerous ways. Her most fundamental critique is that his reliance on switching data as a proxy for diversion is flawed because switching can occur for any number of reasons and, therefore, it is not necessarily indicative of what products consumers would switch to in response to a price increase. DX 17 (Meyer Rep.) at 59-60. Dr. Meyer is certainly correct in this critique. Dr. Warren-Boulton, however, testified forthrightly about the limitations involved in relying on switching data as a proxy for diversion: Using migration [i.e., switching] doesn’t really answer, or it doesn’t answer the precise question of [the] merger guidelines, which of course is, where would you go if there was a small but significant price increase? It basically asks the question, where did you go? And you could go for a lot of reasons. You could go because the price has changed, you could go because the quality changes, you could go because you changed. Complexity changes. And there’s a lot of evidence in the record that people switch because of changes in their own complexity. But using migration percentages, or using those gives you, I think, a reasonable second estimate of diversion ratios, because it’s really asking the question, you know, if you went to some -- if for some reason you decided to go from HRB to TaxACT, for all those reasons, is that roughly about the same percentages if you went due to a price increase? Warren-Boulton, TT, 9/9/11 a.m., at 13-14. Thus, switching data does not necessarily indicate diversion for the reasons both experts have identified. In light of all the evidence in the record and the general similarity of DDIY products, the Court credits Dr. Warren-Boulton’s conclusion 20 Dr. Warren-Boulton’s merger simulation is addressed further below in the Court’s discussion of unilateral effects in Section III.B.2.c. 39 that it was reasonable to use switching data as a proxy for diversion, especially since no more refined historical data apparently exists. Bearing in mind the shortcomings of the switching data, the Court will not treat Dr. Warren-Boulton’s hypothetical monopolist analysis as conclusive. The Court will treat it as another data point suggesting that DDIY is the correct relevant market, however. Another major critique of Dr. Warren-Boulton’s hypothetical monopolist analysis – and one that the defendants repeatedly emphasized at the hearing – is that Dr. Warren-Boulton decided “arbitrarily to exclude some alternatives that are closer substitutes than the products that he included.” DX 17 (Meyer Rep.) at 70; see Meyer, TT, 9/12/11 p.m., at 20-22. As Dr. Meyer put it at the hearing, “Dr. Warren-Boulton’s relevant market is a miscellaneous set of unconnected links, because it doesn’t include . . . the closest substitute to H&R Block [At Home], which is assisted tax preparation. It doesn’t include pen and paper, which is the closest substitute to TaxACT.” Meyer, TT, 9/12/11 p.m., at 24-25. Dr. Meyer identified the “closest substitutes” to the merging parties’ products using simulated diversion data. As discussed below, the Court finds this data unreliable and declines to rely upon it. Dr. Meyer opines, however, that Dr. Warren-Boulton failed to include the closest substitutes for the defendants’ products in his market, even if switching data is treated as a proxy for diversion, as Dr. WarrenBoulton suggests. For example, Dr. Meyer states that “11.2% of TaxACT’s customers in TY2007 switched to assisted preparation in TY2008, while only 2.7% switched to H&R Block At Home and 9.1% switched to TurboTax.” DX 17 (Meyer Rep.) at 72. Thus, the defendants contend Dr. Warren-Boulton violated the following principle from the Merger Guidelines: “When applying the hypothetical monopolist test to define a market around a product offered by one of the merging firms, if the market includes a second product, the Agencies will normally 40 also include a third product if that third product is a closer substitute for the first product than is the second product.” Id. at 72 (quoting Merger Guidelines § 4.1.1). The government persuasively illustrated the key flaw in this critique during the crossexamination of the defendants’ expert. See Meyer, TT, 9/13/11 a.m., at 90-96. Simply put, when determining the “closest substitutes” for products within the DDIY category, Dr. Meyer looked at diversion to individual DDIY brands, such as TurboTax and H&R Block At Home, but when assessing substitutes outside the DDIY category, Dr. Meyer lumped all products and methods together into large, aggregated market categories, such as “assisted” or “pen-andpaper.” See id. If, instead, DDIY products are grouped together as an aggregated category, similar to the treatment of assisted and pen-and-paper in Dr. Meyer’s analysis, then the IRS switching data would indicate that other DDIY products are the closest substitutes for both the DDIY products of HRB and TaxACT. See GTX 15, 16 (illustrating this analysis). For HRB, the numbers show 56.8 percent switching to other DDIY, 36.9 percent to assisted, and 6.3 percent to pen-and-paper. GTX 15. For TaxACT, the numbers show 52.7 percent switching to other DDIY, 40.1 percent to assisted, and 7.3 percent to pen-and-paper. GTX 16. Some of Dr. Meyer’s additional critiques have more merit. For example, one datum Dr. Warren-Boulton relied on in his analysis was the outcome of an advertising study showing that HRB’s sales {were affected} in cities where TaxACT pursued an advertising campaign. See GX 121 (Warren-Boulton Rep.) at 43. The Court accepts Dr. Meyer’s critique that few conclusions can be drawn from this observation because the observed correlation could have been due to other variables – for example, the advertising of a third competitor like TurboTax. See DX 17 (Meyer Rep.) at 69. Similarly, Dr. Warren-Boulton’s observations that changes in relative 41 market share of DDIY, assisted, and pen-and-paper do not appear correlated to changes in relative price could also have been affected by confounding variables. Id. at 67. b. Defendants’ Expert - Dr. Meyer Dr. Meyer found the relevant product market to be all methods of tax preparation, including DDIY, assisted, and pen-and-paper. Her conclusion rested on various factors, including an analysis of documents and testimony. See, e.g., id. at 15. This Court, however, has already discussed its own analysis of the relevant documents and testimony above. Therefore, the Court will focus now on Dr. Meyer’s analysis of pricing data and, in particular, her use of and reliance on data derived from customer surveys commissioned by the defendants. Dr. Meyer found that assisted preparation competes with DDIY in part because the assisted method is the most popular method of tax preparation across all complexity levels. See id. at 12-13. Dr. Meyer concedes, however, that “taxpayers with the most complex tax returns are the most likely to use [assisted preparation].” Id. Indeed, her data show that this effect is pronounced, with approximately 70 percent of filers of complex returns using assisted and approximately 44 percent of filers of simple returns using assisted. Id. DDIY, by contrast, accounts for approximately 37 percent of simple returns and 23 percent of complex returns. Id. If anything, these data indicate that assisted products are linked to the needs of consumers with complex returns, suggesting a partially different consumer profile from DDIY products. Dr. Meyer also noted that the pricing of DDIY and assisted products overlaps, but her analysis of this overlap rests primarily on comparing high-end DDIY products, such as HRB’s Best of Both product,21 with low-end assisted products, such as Jackson Hewitt’s offering of 21 The Best of Both product, as the name implies, actually combines aspects of DDIY and assisted. It enables a return completed on HRB’s DDIY product to be reviewed by a tax professional. See DX 17 (Meyer Rep.) at 13 n.44. Thus, it is hardly surprising that this “hybrid” product, which features such exhaustive service, is priced more expensively than a typical DDIY product. 42 limited, simple return preparation at Wal-Mart. See id. at 13-14. Dr. Meyer concedes that the median price of assisted is higher than the median DDIY price, see id. at 13, and that is the more useful point of comparison. Apart from these comparisons and her conclusions about how industry participants view the market based on her review of documents and testimony in the record, Dr. Meyer’s definition of the relevant market rests primarily on her analysis of simulated diversion data obtained from a “pricing simulator” created for HRB in 2009 and an email survey conducted by TaxACT in 2011. See id. at 17-20. These two sources for her conclusions are discussed seriatim below. i. Pricing Simulator Dr. Meyer asserts in her report that the pricing simulator “created for HRB in 2009, provides the only direct test of the likely diversion from HRB’s [DDIY] products in reaction to a change in price.” Id. at 17. The simulator itself is a pricing model that runs as a dynamic Excel spreadsheet. See Meyer, TT, 9/13/11 a.m., at 42. Dr. Meyer’s report in several instances relies upon an internal HRB Powerpoint presentation that reflects the simulator’s data output under several different scenarios. See, e.g., DX 17 (Meyer Rep.) at 37 n.155 (citing the Powerpoint). As Dr. Meyer describes, the “simulator was prepared using a discrete choice survey of 6,119 respondents.” Id. at 17. She explains that “[t]he respondents were shown five pricing scenarios, and the options included online DIY options, software DIY options, assisted tax preparation options, and other DIY options (including pen-and-paper and friends/family).” Id. Dr. Meyer further states that the “pricing of the various options changed across scenarios” and a “conjoint analysis was conducted to analyze the effect of a change in the price of each product on its own sales and the sales of the other tax preparation options.” Id. 43 Based on the pricing simulator’s results, Dr. Meyer calculated diversion ratios for DDIY products. Dr. Meyer found that “the largest diversion from HRB’s [DDIY products], in the event of a price increase, is to CPAs and accountants.” Id. at 18. She found the “second largest diversion from HRB’s [DDIY products]” was to pen-and-paper. Id. at 19-20. In addition, “the fourth largest diversion is to HRB retail stores.” Id. at 18. Accordingly, Dr. Meyer concluded that assisted preparation and pen-and-paper were the closest substitutes to HRB’s DDIY products and should be included in the relevant market. There is a critical flaw in the design of the pricing simulator, however, that renders conclusions based on its output unreliable. Despite Dr. Meyer’s assertion that the “pricing of the various options changed across [the] scenarios” presented to the survey respondents, not all of the options in the survey underlying the simulator actually had prices associated with them. See Meyer, TT, 9/13/11 a.m., at 27-28. Several “non-priced choice options” were available to the survey respondents and these non-priced options included, importantly, “CPA or Accountant,” “H&R Block Retail Office,” and “Paper & Pencil.” DX 9231 (May 2009 Pricing Simulator Powerpoint) at 4. Thus, while the pricing of the various options changed for some products across the different scenarios presented in the survey, no prices at all were associated with these critical “non-priced choice options.” The fact that the pricing simulator survey failed to assign any prices to these particular products is, of course, especially significant given Dr. Meyer’s findings that the highest diversion from HRB DDIY was to CPAs and then to pen-and-paper. DX 17 (Meyer Rep.) at 18. Indeed, the conclusion that the largest diversion from HRB’s DDIY products would be to CPAs is puzzling on its face. This outcome is counterintuitive because CPAs in general tend to be the most expensive form of tax preparation assistance, while DDIY tends to be the least expensive. 44 See GTX 14. The Court finds that these surprising results are most likely due to the fact that the survey did not, in fact, assign any price at all to the CPA option. Due to this flaw in the survey’s design, respondents may well have selected the CPA option and the other non-priced options without even attempting to consider price as a factor in their decision. Accordingly, the Court finds that it simply cannot rely on the diversion ratios predicted by the simulator. Additional problems with the pricing simulator also render its output unreliable. As Dr. Warren-Boulton noted in his rebuttal of Dr. Meyer’s report, the compilation of pricing simulator data which Dr. Meyer relied upon to calculate her diversion ratios contains results that appear to violate what is “[p]erhaps the most fundamental principle in economics.” See GX 665 (WarrenBoulton Reply Rep.) at 9-10. Increasing the price of one HRB DDIY product in the simulation, TaxCut Online Basic, appears to increase the quantity of the product sold, holding other variables constant. Id. This anomaly violates the fundamental economic principle that “demand curves almost always slope downward,” which holds that, all other things being equal, consumers buy less of a product when the price goes up. See id. In another anomalous result, Dr. Warren-Boulton found that, based on the simulator data, cutting the price of TaxCut Online Basic from $29.95 to $14.95 approximately doubles its predicted market share, but cutting the price only to $19.95 greatly reduces its market share.22 Id. Dr. Warren-Boulton also found that analysis of different print outs of simulator data in the HRB Powerpoint may yield inexplicably different results. For example, relying on the data on one page of the simulator Powerpoint, Dr. Meyer determined that the “the diversion rate from HRB to TaxACT is only 1.6 percent.” DX 17 (Meyer Rep.) at 37. Yet, Dr. Warren-Boulton applied the same methodology for calculating 22 Dr. Meyer testified at the hearing that these anomalies are not reflected in the underlying simulator Excel data, but rather appear only in the printouts of simulator data contained in the internal HRB Powerpoint. In addition, Dr. Meyer explained that she redid her calculations excluding the anomalous data and came up with the same conclusions. See Meyer, TT, 9/12/11 p.m., at 45-47. Dr. Meyer never identified the source or cause of the anomalies, however. Id. at 49. 45 the diversion rate to the simulator data reflected on another slide of the same Powerpoint purporting to show the same simulator data as applied to a different scenario. This calculation yielded the “wildly different estimate” of a 32.4 percent diversion rate from HRB to TaxACT. See GX 665 (Warren-Boulton Reply Rep.) at 10. These inconsistent and anomalous results provide additional reasons to discredit the diversion ratios Dr. Meyer predicted from the simulator data. ii. 2011 Email Survey Dr. Meyer’s analysis also relied on a 2011 email survey of TaxACT customers commissioned by the defendants.23 See DX 17 (Meyer Rep.) at 20, 38. In April 2011, TaxACT and HRB jointly commissioned this survey “to determine to which products TaxACT’s customers would switch if those customers were displeased with TaxACT because of price, quality, or functionality.” Id. at 20. The survey asked one primary question: “If you had become dissatisfied with TaxACT’s price, functionality, or quality, which of these products or services would you have considered using to prepare your federal taxes?” GX 604 (Survey Summary) at 1. The survey then offered the respondents a list of other products or services from which to choose and instructed them to select all applicable options. Id. The list of options that respondents were given varied somewhat depending on the respondents’ filing status and the payments they had made for their 2011 tax returns.24 Id. A follow-up question asked the respondents to narrow their selections to a single choice. Id. 23 Prior to the hearing in this case, the government filed a motion in limine to exclude this survey from evidence and to limit Dr. Meyer’s opinion to the extent it relied on the survey. See ECF No. 60. The government argued that the survey’s wording and methodology made it inherently unreliable and therefore inadmissible. While the Court noted that the government had identified a number of defects in the methodology and wording of the survey, the Court concluded that these defects did not undermine the survey and the expert’s reliance on it so overwhelmingly as to render the survey inadmissible, especially in a bench trial. See Memorandum Opinion and Order on Motion in Limine, September 6, 2011, ECF No. 84. 24 The response options varied among four different categories of filers, which are discussed further below. For example, the list of options presented to filers who completed a free federal tax return and no state return were: “I 46 The research firm conducting the survey initially sent out 46,899 email requests to TaxACT customers inviting them to participate in the survey and then subsequently targeted 24,898 customers who had purchased a federal tax return product but not a state product. Id. Survey respondents were also asked screening questions to determine their membership in one of four categories of customers: (1) those who paid to complete both a federal and state tax return; (2) those who completed a free federal return and paid to complete a state return; (3) those who completed a paid federal return but did not complete a state return; and (4) those who completed a free federal return and did not complete a state return. Id. A total of 1,089 customers responded to the survey. Id. at 1-3. The response rates for the four categories of customers were: (1) 2.45 percent for paid federal / paid state filing (422); (2) 2.08 percent for free federal / paid state filing (245); (3) 0.6 percent for paid federal / no state filing (182); and (4) 1.7 percent for free federal / no state filing (240). Id. Dr. Meyer opined that “this survey is closer to the concept of a diversion ratio than are data on overall switching between products.” DX 17 (Meyer Rep.) at 20 n.85. Based on the survey’s results, she concluded that the survey “provides direct evidence that digital DIY products compete with pen-and-paper” because the percentage of TaxACT customers who reported that, if they were dissatisfied with TaxACT, they would switch to pen-and-paper in each group ranged from 27 to 34 percent. DX 17 (Meyer Rep.) at 20. Dr. Meyer also noted that the survey showed that few TaxACT customers would switch to H&R Block At Home, since only 4 to 10 percent of respondents selected that option. Id. at 38. Accordingly, Dr. Meyer found this outcome indicative that “HRB is not a particularly close competitor to TaxACT.” Id. would prepare myself without help,” “TurboTax Free Edition,” “H&R Block at Home Free Edition,” “Free TaxUSA Free Edition,” “Complete Tax Free Basic,” “An Accountant,” “I would use a product on FFA [i.e., Free File Alliance],” “TaxSlayer Free Edition,” “Jackson Hewitt Free Basic,” “Tax$imple Free Basic,” and “Other.” GX604 at 2. 47 In response to Dr. Meyer’s reliance upon this survey, the government submitted a rebuttal expert report from Dr. Ravi Dhar, a professor of management at Yale University, which credibly critiques the survey on several levels.25 GX 623 (Dhar Rep.). Most fundamentally, the government points out that the phrasing of the survey question – which asks about dissatisfaction with “TaxACT’s price, functionality, or quality” – appears to ask a hypothetical question about switching, not diversion based solely on a price change. Since the phrasing of the survey question conflates customer concerns about price, functionality, or quality, the government argues that the survey cannot shed any light on customer reactions to price changes alone. See id. at 5. Further, to the extent that the wording of the question addresses price, it does not ask about a change in price, but rather suggests a change in the customer’s satisfaction with TaxACT’s existing price. See id. At the hearing, Dr. Meyer explained that she viewed the email survey data as “closer to diversion than is pure switching data” because switching could occur for any reason at all, while the survey only asked about potential switching due to dissatisfaction with “price, functionality, or quality.” Meyer, TT, 9/13/11 a.m., at 87. Yet the Court finds that almost any reason for switching from a product could be characterized as dissatisfaction with the “functionality” or “quality” offered by the product in some respect. Therefore, the survey question does not come much closer to identifying diversion ratios than pure switching data does. Moreover, since there is extensive IRS data reflecting actual switching behavior in the marketplace – as opposed to the hypothetical switching behavior asked about in the email survey – the Court will not rely on the “diversion ratios” suggested by the 2011 email survey. Furthermore, additional defects in the 2011 email survey’s methodology also render the reliability of its findings questionable. First, the high level of non-response to the defendants’ 25 Dr. Dhar did not testify in person at the hearing, but provided an expert report and affidavit. 48 email invitations to participate in the survey could have biased the results. Dr. Dhar explained that the “level of nonresponse . . . is extremely high (more than 98%)” and that the “extremely low response rates makes it difficult to determine whether the results were impacted by a certain segment who were systematically more likely to respond to the survey (e.g., those who were price sensitive or time insensitive) in relation to those who did not respond.” GX 623 (Dhar Rep.) at 10. The Court agrees that non-response bias is a potential pitfall of the survey. See University of Kansas v. Sinks, No. 06-2341, 2008 WL 755065, at *4 (D. Kan. Mar. 19, 2008) (noting, in trademark case, that a consumer survey response rate of “2.16% appears, by any standard, to be quite low.”). Second, by providing survey respondents with a pre-selected list of alternative options, rather than letting respondents respond organically, the survey leads respondents to think about the market for tax preparation services in the same terms that the defendants do, which may have led respondents to select options they otherwise would not have selected. Since the survey’s question essentially asks about hypothetical switching, and since the actual IRS switching data in this case reflect a much larger sample size without the methodological deficiencies of the 2011 survey, the Court declines to rely on the purported diversion ratios calculated from the email survey. On the whole, the Court views Dr. Warren-Boulton’s expert analysis as more persuasive than Dr. Meyer’s.26 First, Dr. Warren-Boulton’s testimony was generally more credible than Dr. Meyer’s.27 Second, the diversion ratios that Dr. Meyer calculated from the pricing simulator and the 2011 email survey are unreliable, as discussed above. Without these simulated diversion 26 Of course, the Court remains cognizant that the plaintiff bears the burden of proof in demonstrating the relevant market. 27 For example, Dr. Meyer’s description of the pricing simulator survey as one in which the “pricing of the various options changed across scenarios” was inaccurate insofar as several of the most significant products for the purposes of Dr. Meyer’s analysis did not have any prices associated with them at all. See discussion supra. 49 ratios, little remains of Dr. Meyer’s expert conclusions apart from her analysis of documents in the record. Dr. Warren-Boulton’s analysis is not without its limitations. The main shortcoming for his approach is that he relied on switching data as a proxy for diversion. Since there is evidence in the record that switching among different products in the broader tax preparation industry occurs for reasons other than price competition, switching cannot serve as a complete proxy for diversion. Even so, the Court credits Dr. Warren-Boulton’s conclusion that switching data can provide a “reasonable second estimate” of diversion ratios here. Therefore, the Court finds that Dr. Warren-Boulton’s analysis tends to confirm that the relevant market is DDIY, although the Court would not rely on his analysis exclusively. As explained above, however, the full body of evidence in this case makes clear that DDIY is the correct relevant market for evaluating this merger. B. Likely Effect on Competition 1. The Plaintiff’s Prima Facie Case Having defined the relevant market as DDIY tax preparation products, “the Court must next consider the likely effects of the proposed acquisition on competition within that market.” Swedish Match, 131 F. Supp. 2d at 166. The government must now make out its prima facie case by showing “that the merger would produce ‘a firm controlling an undue percentage share of the relevant market, and [would] result[ ] in a significant increase in the concentration of firms in that market.’” Heinz, 246 F.3d at 715 (quoting Philadelphia Nat’l Bank, 374 U.S. at 363). “Such a showing establishes a ‘presumption’ that the merger will substantially lessen competition.” Id. 50 “Market concentration, or the lack thereof, is often measured by the HerfindahlHirschmann Index (‘HHI’).” Id. at 716. “The HHI is calculated by totaling the squares of the market shares of every firm in the relevant market. For example, a market with ten firms having market shares of 20%, 17%, 13%, 12%, 10%, 10%, 8%, 5%, 3% and 2% has an HHI of 1304 (202 + 172 + 132 + 122 + 102 + 102 + 82 + 52 +32 +22).” Id. at 715 n.9. Sufficiently large HHI figures establish the government’s prima facie case that a merger is anticompetitive. Id. Under the Horizontal Merger Guidelines, markets with an HHI above 2500 are considered “highly concentrated” and mergers “resulting in highly concentrated markets that involve an increase in the HHI of more than 200 points will be presumed to be likely to enhance market power.” Merger Guidelines § 5.3. In this case, market concentration as measured by HHI is currently 4,291, indicating a highly concentrated market under the Merger Guidelines. GX 121 (Warren-Boulton Rep.) at 38. The most recent measures of market share show Intuit with 62.2 percent of the market, HRB with 15.6 percent, and TaxACT with 12.8 percent. GX 27. These market share calculations are based on data provided by the IRS for federal tax filings for 2010, the most recent data available. The defendants argue that market share calculations based exclusively on federal filing data are insufficient to meet the plaintiff’s burden in establishing its alleged relevant product market, which includes both federal and state filings. Defs.’ Post-Trial Mem. at 12-13. The Court rejects this argument. State tax return products are typically sold as add-ons to or in combination with federal return products and the Court finds that there is little reason to conclude that the market share proportions within the state DDIY segment would be significantly different from federal DDIY. See GX 600 at 8 (HRB market research study stating that “[t]he desire to file State and Federal taxes together, and, inherently, for ease/convenience overruled all 51 other rationales for the method chosen for State taxes.”). While, as defendants point out, many customers of federal tax return DDIY products do not also purchase state returns, that may be because they live in states without income tax or because their state returns are simple enough to prepare very easily without assistance. See Dunn, TT, 9/8/11 a.m., at 48-49. A reliable, reasonable, close approximation of relevant market share data is sufficient, however. FTC v. PPG Indus., Inc., 798 F.2d 1500, 1505 (D.C. Cir. 1986). Further, the defendants’ own ordinary course of business documents analyze the market based on IRS federal e-file data, without reference to state filings, even though the defendants’ clearly sell state tax return products. See, e.g., GX 27. The proposed acquisition in this case would give the combined firm a 28.4 percent market share and will increase the HHI by approximately 400, resulting in a post-acquisition HHI of 4,691. Id. These HHI levels are high enough to create a presumption of anticompetitive effects. See, e.g., Heinz, 246 F.3d at 716 (three-firm to two-firm merger that would have increased HHI by 510 points from 4,775 created presumption of anticompetitive effects by a “wide margin”); Swedish Match, 131 F. Supp. 2d at 166-67 (60 percent market share and 4,733 HHI established presumption). Accordingly, the government has established a prima facie case of anticompetitive effects. “Upon the showing of a prima facie case, the burden shifts to defendants to show that traditional economic theories of the competitive effects of market concentration are not an accurate indicator of the merger’s probable effect on competition in these markets or that the procompetitive effects of the merger are likely to outweigh any potential anticompetitive effects.” CCC Holdings, 605 F. Supp. 2d at 46. “The courts have not established a clear standard that the merging parties must meet in order to rebut a prima facie case, other than to 52 advise that ‘[t]he more compelling the prima facie case, the more evidence the defendant must present to rebut [the presumption] successfully.’” Id. at 46-47 (quoting Baker Hughes, 908 F.2d at 991). Even in cases where the government has made a strong prima facie showing: [i]mposing a heavy burden of production on a defendant would be particularly anomalous where, as here, it is easy to establish a prima facie case. The government, after all, can carry its initial burden of production simply by presenting market concentration statistics. To allow the government virtually to rest its case at that point, leaving the defendant to prove the core of the dispute, would grossly inflate the role of statistics in actions brought under section 7. The Herfindahl-Hirschman Index cannot guarantee litigation victories. Baker Hughes, 908 F.2d at 992. Thus, ultimately,“[t]he Supreme Court has adopted a totality-ofthe-circumstances approach to the [Clayton Act], weighing a variety of factors to determine the effects of particular transactions on competition.” Id. at 984. With these observations in mind, the Court will evaluate the parties’ evidence and arguments about the likely effect of the transaction on competition in the DDIY market. 2. Defendants’ Rebuttal Arguments a. Barriers to Entry Defendants argue that the likelihood of expansion by existing DDIY companies besides Intuit, HRB, and TaxACT will offset any potential anticompetitive effects from the merger. Courts have held that likely entry or expansion by other competitors can counteract anticompetitive effects that would otherwise be expected. See Heinz, 246 F.3d at 717 n.13 (“Barriers to entry are important in evaluating whether market concentration statistics accurately reflect the pre- and likely post-merger competitive picture.”); Baker Hughes, 908 F.2d at 987 (“In the absence of significant barriers, a company probably cannot maintain supracompetitive pricing for any length of time.”). According to the Merger Guidelines, entry or expansion must be “timely, likely, and sufficient in its magnitude, character, and scope to deter or counteract the competitive effects of concern.” Merger Guidelines § 9; see also CCC Holdings, 605 F. Supp. 53 2d at 47; United States v. Visa USA, Inc., 163 F. Supp. 2d 322, 342 (S.D.N.Y. 2001) (entry must be “timely, likely, and [of a] sufficient scale to deter or counteract any anticompetitive restraints”). “Determining whether there is ease of entry hinges upon an analysis of barriers to new firms entering the market or existing firms expanding into new regions of the market.” CCC Holdings, 605 F. Supp. 2d at 47 (quoting FTC v. Cardinal Health, Inc., 12 F. Supp. 2d 34, 55 (D.D.C. 1998)). In this case, the parties essentially agree that the proper focus of this inquiry is on the likelihood of expansion by existing competitors rather than new entry into the market.28 See Defs.’ Post-Trial Mem. at 21-22. Since the government has established its prima facie case, the defendants carry the burden to show that ease of expansion is sufficient “to fill the competitive void that will result if [defendants are] permitted to purchase” their acquisition target. Swedish Match, 131 F. Supp. 2d at 169. In describing the competitive landscape, the defendants note there are eighteen companies offering various DDIY products through the FFA. Defs.’ Post-Trial Mem. at 22. Most of these companies are very small-time operators, however. The defendants acknowledge this fact, but nevertheless contend that the companies “TaxSlayer and TaxHawk are the two largest and most poised to replicate the scale and strength of TaxACT.” Id. at 23. Witnesses from TaxSlayer and TaxHawk were the only witnesses from other DDIY companies to testify at the hearing. As such, the Court’s ease of expansion analysis will focus on whether these two competitors are poised to expand in a way that is “timely, likely, and sufficient in its magnitude, 28 New entrants to the market would not only face all of the barriers to expansion already faced by the existing small firms offering DDIY products, they would also have to develop their own products, including a software platform and a sufficient level of tax expertise. For entry to be considered timely, it typically must occur within approximately two years post-merger. See Commentary on the Horizontal Merger Guidelines (2006) at 45-46 (discussing prior Merger Guidelines § 3.2, which specified that timely entry should occur within two years). It is unlikely that an entirely new entrant to the market could compete meaningfully with the established DDIY firms within that time frame. 54 character, and scope to deter or counteract” any potential anticompetitive effects resulting from the merger. TaxHawk runs five different websites, including FreeTaxUSA.com, that all market the same underlying DDIY product. Kimber, TT, 9/12/11 a.m., at 12, 40. TaxHawk was founded in 2001, three years after TaxACT, although it has a significantly smaller market share of 3.2 percent. Id. at 11; GX 27. TaxHawk’s vice-president and co-founder, Mr. Dane Kimber, testified that the company has the technical infrastructure to grow by five to seven times the number of customers in any given year. Kimber, TT, 9/12/11 a.m., at 21. TaxHawk’s marketing strategy relies substantially on search engine advertising and search term optimization, including by using the FreeTaxUSA.com domain name, which contains the keywords “free” and “tax.” See id. at 19-27. Despite having been in business for a decade, its products are functionally more limited than those of Intuit, HRB, and TaxACT in various ways. See PFF ¶ 185. Although TaxHawk services the forms that cover most taxpayers, its program does not service all federal forms, it excludes two states’ forms in their entirety, and it does not service city income tax forms for major cities that have income taxes – notably, New York City. Kimber, TT, 9/12/11 a.m., at 44. In fact, Mr. Kimber testified that the company would likely need another decade before its DDIY products could fully support all the tax forms. Id. at 45. The reason is that TaxHawk is what Mr. Kimber “like[s] to call . . . a ‘lifestyle’ company. We like the lifestyle we have as owners. We want our employees to have a life, if you will. I do feel we have the expertise to [expand functionality] more rapidly, but we choose not to.” Id. Mr. Kimber also testified that TaxHawk had suddenly experienced an unprecedented growth rate of over 60 percent since April 2011, id. at 20-21, but that the company had not done any analysis to attempt to explain this unanticipated (and presumably welcome) growth. Id. at 39. 55 TaxHawk’s relaxed attitude toward its business stands in stark contrast to the entrepreneurial verve that was apparent throughout the testimony of Mr. Dunn and that has been rewarded by the impressive growth of TaxACT over the years. In short, TaxHawk is a very different company from TaxACT. TaxHawk is a small company that has developed a string of search-engine-optimized DDIY websites, which deliver a sufficient income stream to sustain its owners’ comfortable lifestyle, without requiring maximal effort on their part. While TaxHawk’s decision to prioritize a relaxed lifestyle over robust competition and innovation is certainly a valid one, expansion from TaxHawk that would allow it to compete “on the same playing field” as the merged company appears unlikely. Chicago Bridge & Iron Co. N.V. v. FTC, 534 F.3d 410, 430 (5th Cir. 2008). After TaxHawk, TaxSlayer is the next largest DDIY competitor, with a 2.7 percent market share. GX 27. TaxSlayer.com launched in 2003, although the same company started selling a software product to tax professionals several years earlier. Rhodes, TT, 9/12/11 a.m., at 71. TaxSlayer is part of the same corporate family as Rhodes Murphy, a tax firm that provides assisted tax preparation through sixteen retail offices in the Augusta, Georgia area. Id. The company is a family business and James Brian Rhodes, the product manager of TaxSlayer and the son of the company’s founder, testified at the hearing. Id. at 70,73. Mr. Rhodes testified that, in the event of an increase in TaxACT’s prices or a decrease in its quality, he “believe[s] that [TaxSlayer is] poised and ready to take those customers who would want to go elsewhere for lower prices.” Id. at 81. TaxSlayer’s marketing strategy relies heavily on sponsorship of sporting events, including the Gator Bowl and NASCAR. Id. at 75. TaxSlayer typically invests {a significant amount of its budget in marketing}. Rhodes, TT, 9/12/11 a.m. (sealed), at 86, 92. For example, TaxSlayer plans to spend ${redacted} on marketing in 2012 based on 2011 56 revenues of ${redacted}. Id. at 84, 87. Despite this {high} level of marketing spending, TaxSlayer’s DDIY market share has not changed substantially since 2006, despite steady growth in TaxSlayer’s revenue and number of units sold. See id. at 94-96; GX 21-7 (IRS e-file share data chart showing 2.5 percent share for TaxSlayer in 2006 and 2.7 percent in 2010). Rather, TaxSlayer’s growth in unit sales and revenue has come from maintaining the same slice of an expanding pie – the growing DDIY market. See GX 21-7. TaxSlayer’s stable market share despite its {significant} marketing expenditure as a proportion of revenue points to what the government considers the key barrier to entry in this market – the importance of reputation and brand in driving consumer behavior in purchasing DDIY products. Simply put, tax returns are highly personal documents that carry significant financial and legal consequences for consumers. Consumers, therefore, must trust and have confidence in their tax service provider. As one of TaxACT’s bankers put it a confidential memorandum, “[t]ax filers must have confidence that sensitive data is being handled with care and that returns are processed in a secure, error-free and timely manner.” GX 125 at 12. Building a reputation that a significant number of consumers will trust requires time and money. As HRB’s former CEO noted, it takes millions of dollars and lots of time to develop a brand. Bennett, TT, 9/6/11 p.m., at 30. TaxACT’s offering memoranda also point to the difficulty in building a brand in the industry as a barrier to competition. See GX 28-24 at 2SSCORPe-2419 (2009 memorandum stating “With over 11 years of building reliable, robust software solutions, 2SS has created a valuable brand within the online tax preparation market which Management believes would take years of competitive investment to replicate.”). In the DDIY industry, the Big Three incumbent players spend millions on marketing and advertising each year to build and maintain their brands, dwarfing the combined spending of the smaller 57 companies. For example, in tax year 2009, Intuit, HRB, and TaxACT collectively spent approximately {over $100 million} on marketing and advertising. GX 29 (Intuit Decl.) ¶ 38; GX 61-22 at 3; GX 138 at 37. By contrast, {TaxSlayer and TaxHawk spent a significantly smaller amount}.29 Rhodes, TT, 9/12/11 a.m. (sealed), at 95; GX 25 (TaxHawk Decl.) ¶ 14. Even TaxACT’s successful business strategy has been premised on the notion that it cannot outspend Intuit and HRB on marketing. Dunn, TT, 9/7/11 p.m., at 71-72. The massive marketing expenditures of the two major DDIY firms create high per customer acquisition costs and limit the easy marketing channels that are open to smaller competitors. See, e.g., id. at 88-89 (noting that “Web advertising is the most competitive. . . I think [TaxACT is] going to get shut out on Yahoo [the popular web portal]. I think Intuit is going to buy it lock, stock and barrel,” and explaining that this outcome would hurt TaxACT’s business if it doesn’t find effective alternative advertising venues). Rather than attempting to outspend HRB and Intuit, TaxACT’s growth strategy has largely depended on providing “great customer service, a great product, and a great customer experience” and then relying on word-of-mouth referrals to spread the awareness of the brand. Id. at 71-72. This process is inherently time-consuming and difficult to replicate. In support of their argument that TaxSlayer and TaxHawk are poised to expand in response to a price increase, the defendants emphasize that these companies “are at about the same position in terms of customer base as TaxACT was in 2002, which was the year before it did the Free For All [offer on] the FFA.” Meyer, TT, 9/12/11 p.m., at 130. The government points out, however, that there are two flaws in this comparison, even assuming that TaxSlayer 29 The defendants attempt to reframe this disparity by noting that their calculation of TaxSlayer’s projected tax season 2015 marketing budget would slightly surpass the amount of TaxACT’s actual 2011 marketing budget. Defs.’ Post-Trial Mem. at 23. Setting aside the validity of the defendants’ aggressive projections of TaxSlayer’s 2015 budget, a proper comparison would have to be founded upon a comparable projection of TaxACT’s 2015 budget—not TaxACT’s actual 2011 numbers, for which the relevant comparison is TaxSlayer’s 2011 numbers. 58 and TaxHawk were TaxACT’s competitive equals. First, while these companies may have a similar number of customers to TaxACT in 2002 in absolute terms, TaxACT’s market share at 8 percent was already significantly larger than the market shares of these firms today, despite the fact that TaxACT had been in the market for fewer years. See GTX 17. Second, the DDIY market has matured considerably since 2002, in parallel with the general ripening of various online industries during the past decade. Notably, the pool of penand-paper customers has dwindled as DDIY preparation has grown. Thus, the “low hanging fruit” of DDIY customer acquisition may have been plucked. See GX 296 (Houseworth Dep.) at 66-68 (noting that “there’s probably only two or three years of continued mid teens category growth for online” because of the shrinking pool of new potential customers that can be converted from the pen-and-paper method). This trend suggests existing market shares may become further entrenched and that growing market share may be even harder, especially because there are barriers to switching from one DDIY product to another. For example, the hearing evidence showed that it is difficult to import prior-year tax return data across DDIY brands. If a taxpayer uses, say, TurboTax or TaxACT in one year, then when the taxpayer returns the next year, the program can automatically import the prior year’s data, which is not only convenient but can also help the taxpayer identify useful tax information, such as carry forwards and available deductions. Dunn, TT, 9/8/11 a.m., at 111-14. Currently, it is not possible to import much of this data if the taxpayer switches to a competitor’s product. Id. Thus, this feature lends a “stickiness” to each particular DDIY product once a customer has used it. Upon consideration of all of the evidence relating to barriers to entry or expansion, the Court cannot find that expansion is likely to avert anticompetitive effects from the transaction. 59 The Court will next consider whether the evidence supports a likelihood of coordinated or unilateral anticompetitive effects from the merger. b. Coordinated Effects Merger law “rests upon the theory that, where rivals are few, firms will be able to coordinate their behavior, either by overt collusion or implicit understanding in order to restrict output and achieve profits above competitive levels.” CCC Holdings, 605 F. Supp. 2d at 60 (quoting Heinz, 246 F.3d at 715). The government argues that the “elimination of TaxACT, one of the ‘Big 3’ Digital DIY firms” will facilitate tacit coordination between Intuit and HRB. Pl.’s Post-Trial Mem. at 15. “Whether a merger will make coordinated interaction more likely depends on whether market conditions, on the whole, are conducive to reaching terms of coordination and detecting and punishing deviations from those terms.” CCC Holdings, 605 F. Supp. 2d at 60 (internal quotation omitted). Since the government has established its prima facie case, the burden is on the defendants to produce evidence of “structural market barriers to collusion” specific to this industry that would defeat the “ordinary presumption of collusion” that attaches to a merger in a highly concentrated market. See Heinz, 246 F.3d at 725. The defendants argue the primary reason that coordinated effects will be unlikely is that Intuit will have no incentive to compete any less vigorously post-merger. The defendants assert that the competition between Intuit and HRB’s retail stores would be “fundamentally nullified if Intuit decided to reduce the competitiveness of TurboTax.” Defs.’ Post-Trial Mem. at 17. Further, defendants contend that Intuit has no incentive to reduce the competitiveness of its free product because it views its free product as a critical driver of new customers. Id. at 17-18 Therefore, the defendants conclude that if HRB does not compete as aggressively as possible with its post-merger products, it will lose customers to Intuit. Id. at 18. 60 The most compelling evidence the defendants marshal in support of these arguments consists of documents and testimony indicating that Intuit engaged in a series of “war games” designed to anticipate and defuse new competitive threats that might emerge from HRB postmerger. See GX 293 (Intuit Dep.) at 98-101; DX 84. The documents and testimony do indicate that Intuit and HRB will continue to compete for taxpayers’ patronage after the merger—indeed, in the DDIY market, they would be the only major competitors. This conclusion, however, is not necessarily inconsistent with some coordination. As the Merger Guidelines explain, coordinated interaction involves a range of conduct, including unspoken understandings about how firms will compete or refrain from competing. See Merger Guidelines § 7. In this case, the government contends that coordination would likely take the form of mutual recognition that neither firm has an interest in an overall “race to free” in which highquality tax preparation software is provided for free or very low prices. Indeed, the government points to an outline created as part of the Intuit “war games” regarding post-merger competition with HRB that also indicates an Intuit employee’s perception that part of HRB’s post-merger strategy would be to “not escalate free war: Make free the starting point not the end point for customers.” GX 293-13 at INT-DOJ0015942.30 Since, as defendants point out, DDIY companies have found “free” offers to be a useful marketing tool, it is unlikely that free offers would be eliminated. Rather, the government argues, it is more likely that HRB and Intuit may 30 The government also cites an informal analysis written by Adam Newkirk, an analyst for HRB’s DDIY business. Mr. Newkirk’s analysis hypothesized that one possible reason for HRB to acquire TaxACT was that HRB and Intuit would jointly control a large DDIY market share post-merger and would “both obviously have great incentive to keep this channel profitable,” while other potential purchasers of TaxACT “could decide to cut prices even further . . . .” See Newkirk, TT, 9/7/11 a.m., at 100; GX 18. The Court finds that the government overemphasized the importance and relevance of Mr. Newkirk’s analysis. The hearing testimony showed that Mr. Newkirk is a data analyst who had no decision-making role or authority in relation to the merger and that his discussion about the rationales for the merger was informal speculation. See Newkirk, TT, 9/7/11 p.m., at 42-44. Even so, this reasoning – independently reached by Intuit – is essentially a précis of the government’s coordinated effects concern. 61 find it “in their mutual interest to reduce the quality of their free offerings . . . offer a lower quality free product and maintain higher prices for paid products . . . .” PFF ¶ 141. The government points to a highly persuasive historical act of cooperation between HRB and Intuit that supports this theory. Cf. Merger Guidelines § 7.2 (“[M]arket conditions are conducive to coordinated interaction if firms representing a substantial share in the relevant market appear to have previously engaged in express collusion.”). After TaxACT launched its free-for-all offer in the FFA, Intuit proposed that the firms in the market limit their free FFA offers, a move which TaxACT opposed and which Mr. Dunn believed was an illegal restraint on trade. Dunn, TT, 9/7/11 p.m., at 79. HRB, Intuit, and others then joined together and successfully lobbied the IRS for limitations on the scope of the free offers through the FFA – limitations that remain in place today. Ernst, TT, 9/7/11 a.m., at 26-27; Warren-Boulton, TT, 9/9/11 p.m., at 78. This action illustrates how the pricing incentives of HRB and Intuit differ from those of TaxACT and it also shows that HRB and Intuit, although otherwise competitors, are capable of acting in concert to protect their common interests. The defendants also argue that coordinated effects are unlikely because the DDIY market consists of differentiated products and has low price transparency. See CCC Holdings, 605 F. Supp. 2d at 62 (recognizing the importance of price transparency to the likelihood of coordinated effects). To the contrary, the record clearly demonstrates that the players in the DDIY industry are well aware of the prices and features offered by competitors. Since DDIY products are marketed to a large swath of the American population and available via the Internet, DDIY firms can easily monitor their competitors’ offerings and pricing. The fact that competitors may offer various discounts and coupons to some customers via email hardly renders industry pricing “not transparent,” as defendants submit. See Defs.’ Post-Trial Mem. at 21. Moreover, while 62 collusion may, in some instances, be more likely in markets for homogenous products than differentiated products, product differentiation in this market would not necessarily make collusion more difficult. See Heinz, 246 F.3d at 716-17, 724-25 (finding likelihood of coordinated effects in product market differentiated by brand); see also CCC Holdings, 605 F. Supp. 2d at 65 n.42 (“[T]acit collusion may be easier when products are differentiated.”) (quoting Lawrence A. Sullivan & Warren S. Grimes, The Law of Antitrust: An Integrated Handbook, § 11.2e1, at 635 (2d ed. 2006)). Other indicia of likely coordination are also present in the DDIY market. Transactions in the market are small, numerous, and spread among a mass of individual consumers, each of whom has low bargaining power; prices can be changed easily; and there are barriers to switching due to the “stickiness” of the DDIY products. See CCC Holdings, 605 F. Supp. 2d at 65-66 (discussing these factors as characteristic of markets conducive to coordination); see also supra Section III.B.2.a (discussing the difficulty of importing data as a barrier to switching from one DDIY product to another). Finally, the Court notes that the “merger would result in the elimination of a particularly aggressive competitor in a highly concentrated market, a factor which is certainly an important consideration when analyzing possible anti-competitive effects.” Staples, 970 F. Supp. at 1083; see also FTC v. Libbey, 211 F. Supp. 2d 34, 47 (D.D.C. 2002). The evidence presented at the hearing from all parties demonstrated TaxACT’s impressive history of innovation and competition in the DDIY market. Mr. Dunn’s trial testimony revealed him to be a dedicated and talented entrepreneur and businessman, with deep knowledge and passion for providing highquality, low-cost tax solutions. TaxACT’s history of expanding the scope of its high-quality, free product offerings has pushed the industry toward lower pricing, even when the two major 63 players were not yet ready to follow – most notably in TaxACT’s introduction of free-for-all into the market. The government presses the argument that TaxACT’s role as an aggressive competitor is particularly important by urging this Court to find that TaxACT is a “maverick.” See Pl.’s PostTrial Mem. at 18-19. In the context of antitrust law, a maverick has been defined as a particularly aggressive competitor that “plays a disruptive role in the market to the benefit of customers.” Merger Guidelines § 2.1.5. The most recent revision of the Merger Guidelines endorses this concept and gives a few examples of firms that may be industry mavericks, such as where “one of the merging firms may have the incentive to take the lead in price cutting or . . . a firm that has often resisted otherwise prevailing industry norms to cooperate on price setting or other terms of competition.” Id. The parties have spilled substantial ink debating TaxACT’s maverick status. The arguments over whether TaxACT is or is not a “maverick” – or whether perhaps it once was a maverick but has not been a maverick recently – have not been particularly helpful to the Court’s analysis. The government even put forward as supposed evidence a TaxACT promotional press release in which the company described itself as a “maverick.” See GX 28-6. This type of evidence amounts to little more than a game of semantic gotcha. Here, the record is clear that while TaxACT has been an aggressive and innovative competitor in the market, as defendants admit, TaxACT is not unique in this role. Other competitors, including HRB and Intuit, have also been aggressive and innovative in forcing companies in the DDIY market to respond to new product offerings to the benefit of consumers. See Defs.’ Post-Trial Mem. at 20. The government has not set out a clear standard, based on functional or economic considerations, to distinguish a maverick from any other aggressive competitor. At times, the 64 government has emphasized TaxACT’s low pricing as evidence of its maverick status, while, at other times, the government seems to suggest that almost any competitive activity on TaxACT’s part is a “disruptive” indicator of a maverick. For example, the government claims that “[m]ost recently, TaxACT continued to disrupt the Digital DIY market by entering the boxed retail software segment of the market, which had belonged solely to HRB and [Intuit].” Pl.’s PostTrial Mem. at 19. Credible evidence at the hearing, however, showed {otherwise}. See Dunn, TT, 9/8/11 p.m. (sealed), at 4. Moreover, the Court credits Mr. Dunn’s explanation that TaxACT has little interest in selling boxed retail software because he believes this market segment is {redacted} not particularly significant. See Dunn, TT, 9/7/11 p.m. (sealed), at 123 ({redacted}). What the Court finds particularly germane for the “maverick” or “particularly aggressive competitor” analysis in this case is this question: Does TaxACT consistently play a role within the competitive structure of this market that constrains prices? See Staples, 970 F. Supp. 1083 (finding “merger would result in the elimination of a particularly aggressive competitor in a highly concentrated market” where the merger would remove competition between “the two lowest cost and lowest priced firms” in the market); Merger Guidelines § 2.1.5 (noting maverick concerns may arise where “one of the merging firms may have the incentive to take the lead in price cutting or [with] . . . a firm that has often resisted otherwise prevailing industry norms to cooperate on price setting or other terms of competition.”). The Court finds that TaxACT’s competition does play a special role in this market that constrains prices. Not only did TaxACT buck prevailing pricing norms by introducing the free-for-all offer, which others later matched, it has remained the only competitor with significant market share to embrace a business strategy that relies primarily on offering high-quality, full-featured products for free with associated products at low prices. 65 Moreover, as the plaintiff’s expert, Dr. Warren-Boulton, explained, the pricing incentives of the merged firm will differ from those of TaxACT pre-merger because the merged firm’s opportunity cost for offering free or very low-priced products will increase as compared to TaxACT now. See Warren-Boulton, 9/9/11 p.m., at 14-16. In other words, the merged firm will have a greater incentive to migrate customers into its higher-priced offerings – for example, by limiting the breadth of features available in the free or low-priced offerings or only offering innovative new features in the higher-priced products. See Commentary on the Horizontal Merger Guidelines (2006) at 24 (noting the importance of asking “whether the acquired firm has behaved as a maverick and whether the incentives that are expected to guide the merged firm’s behavior likely would be different.”). While the defendants oppose the government’s maverick theory, they do not deny that TaxACT has been an aggressive competitor. Indeed, they submit that “that’s why H&R Block wants to buy them.” Defs.’ Closing Argument, TT, 10/3/11 a.m., at 132. HRB contends that the acquisition of TaxACT will result in efficiencies and management improvements that “will lead to better, more effective, and/or cheaper H&R Block digital products post-merger” that are better able to compete with Intuit. Defs.’ Post-Trial Mem. at 17. This argument is quite similar to the argument of the defendants in Heinz, which some commentators have described as arguing that the merger would create a maverick. Heinz, 246 F.3d at 720-22; see Jonathan B. Baker, Mavericks, Mergers, and Exclusion: Proving Coordinated Competitive Effects Under the Antitrust Laws, 77 N.Y.U. L. Rev. 135, 184 (2002). While the district court in Heinz accepted this argument that the merger would enhance rather than stifle competition, the D.C. Circuit reversed, finding that the “district court’s analysis [fell] short of the findings necessary for a successful efficiencies defense” in that case. Heinz, 246 F.3d at 721. As explained more fully in 66 Section III.B.2.d below, the defendants’ efficiency arguments fail here for some of the same reasons the D.C. Circuit identified in Heinz. Finally, the defendants suggest that coordinated effects are unlikely because of the ease of expansion for other competitors in the market. As detailed above in the Court’s discussion of barriers to entry and expansion, the Court does not find that ease of expansion would counteract likely anticompetitive effects. Accordingly, the defendants have not rebutted the presumption that anticompetitive coordinated effects would result from the merger. To the contrary, the preponderance of the evidence suggests the acquisition is reasonably likely to cause such effects. See id. at 711-12 (finding, in market characterized by high barriers to entry and high HHI figures, that “no court has ever approved a merger to duopoly under similar circumstances.”). c. Unilateral Effects A merger is likely to have unilateral anticompetitive effect if the acquiring firm will have the incentive to raise prices or reduce quality after the acquisition, independent of competitive responses from other firms. See Swedish Match, 131 F. Supp. 2d at 169; Merger Guidelines § 6 (“The elimination of competition between two firms that results from their merger may alone constitute a substantial lessening of competition.”). “The extent of direct competition between the products sold by the merging parties is central to the evaluation of unilateral price effects.” Merger Guidelines § 6.1. As Judge Collyer in CCC Holdings explained: Unilateral effects in a differentiated product market are likely to be profitable under the following conditions: (1) the products must be differentiated; (2) the products controlled by the merging firms must be close substitutes, i.e., “a substantial number of the customers of one firm would turn to the other in response to a price increase”; (3) other products must be sufficiently different from the products offered by the merging firms that a merger would make a small but significant and non-transitory price increase profitable for the merging firm; and (4) repositioning must be unlikely. 67 605 F. Supp. 2d at 68 (citing Oracle, 331 F. Supp. 2d at 1117-18).31 Since the Court has already found that the preponderance of the evidence shows a reasonable likelihood of coordinated effects, the Court need not reach the issue of unilateral effects. See id. at 67. The Court will discuss it, however, since there has been substantial argument on this topic and the Court’s findings regarding unilateral effects bolster the conclusion that this proposed merger would violate Section 7 of the Clayton Act. As with coordinated effects, since the government has established its prima facie case, the burden is on the defendants to produce evidence showing that the presumption of anticompetitive effects that attaches to a merger in a highly concentrated market is unfounded, but the ultimate burden of proof remains with the government. i. Elimination of Direct Competition Between the Merging Parties The government argues that unilateral effects are likely because the merger will eliminate head-to-head competition between HRB and TaxACT that has benefited taxpaying American consumers. Much of the evidence indicating direct competition between HRB and TaxACT is discussed above in relation to the market definition. See supra Section III.A. The government emphasizes that HRB has lowered its DDIY prices to better compete with free online products, the category pioneered by TaxACT, and has directly considered TaxACT’s prices in setting its own prices. See GX 53 at 2, 8; GX 188; GX 199 at 5-9. HRB has also determined the nature of its free offerings in response to competitive activity from TaxACT. See, e.g., GX 304 at 5 (HRB changed timing of FFA offering in response to TaxACT’s offer); GX 44 (recognizing need to compete with TaxACT offerings); GX 79 (comparing contemplated free product description on HRB’s website with TaxACT’s website); GX 51 at 4 (noting launch of free online products intended “[t]o match competitor offerings and stem online share loss to Intuit and TaxACT”). The government also points to HRB documents that appear to acknowledge that TaxACT has put 31 The first criterion in this analysis is satisfied because it is undisputed that DDIY products are differentiated. 68 downward pressure on HRB’s pricing ability. See GX 296-16 at 20-21 (noting TaxACT’s association with the “commoditization of online space” and downward price pressure from commoditization); GX 20 at 11 ({redacted}). From all of this evidence, and the additional evidence discussed in this opinion, it is clear that HRB and TaxACT are head-to-head competitors. ii. Pledge to Maintain TaxACT’s Current Prices Defendants press a few different arguments against a finding of likely unilateral anticompetitive effects. First, the defendants have pledged to maintain TaxACT’s current prices for three years.32 While the Court has no reason to doubt that defendants would honor their promise, this type of guarantee cannot rebut a likelihood of anticompetitive effects in this case. See Cardinal Health, 12 F. Supp. 2d at 64 (finding that “even with such guarantees [to maintain prices], the mergers would likely result in anti-competitive prices.”). Even if TaxACT’s list price remains the same, the merged firm could accomplish what amounts to a price increase through other means. For example, instead of raising TaxACT’s prices, it could limit the functionality of TaxACT’s products, reserving special features or innovations for higher priced, HRB-branded products. The merged firm could also limit the availability of TaxACT to consumers by marketing it more selectively and less vigorously. Indeed, the defendants concede that one immediate effect of the merger will be the removal of TaxACT from the IRS-sponsored FFA website, a marketing channel whose importance the defendants themselves emphasize in their argument regarding barriers to expansion. See Dunn, TT, 9/7/11 p.m., at 76-77; Defs.’ Post-Trial Mem. at 22. 32 Before the hearing, the plaintiff filed a motion in limine to exclude evidence relating to this guarantee. ECF No. 44. Following oral argument at the pre-hearing conference, the plaintiff withdrew this motion. See Minute Entry dated September 2, 2011. 69 iii. Value Versus Premium Market Segments Second, defendants argue that HRB and TaxACT are not particularly close competitors. The defendants contend that HRB and TaxACT largely compete in distinct segments of the market – with HRB in the higher-priced, “premium” segment and TaxACT in the lower-priced, “value” segment.33 The defendants also argue that there can be no unilateral effects because the evidence shows that both TaxACT and HRB are closer competitors to TurboTax than to each other. Defs.’ Post-Trial Mem. at 15. As part of the argument that HRB and TaxACT focus on separate value and premium segments, the defendants argued that for several years in the mid-2000s, HRB was trapped in the “murky middle” between TaxACT’s value offerings and Intuit’s premium offerings. See DX 17 (Meyer Rep.) at 29; Meyer, TT, 9/13/2011 a.m., at 103-107. The defendants argue that, in recent years, HRB has positioned itself more clearly as a premium provider, as evidenced by the fact that the list price of its online federal plus state DDIY product has tracked Intuit’s price more closely since 2010. See DX 17 (Meyer Rep.) at 29. This comparison is misleading because it focuses solely on the comparison of the list prices for the companies’ highest-priced products. See id. at 29 n.116. During the past few years, while HRB has increased the list price of its toppriced DDIY offering, it has also more heavily marketed free products. See GX 51 at 4; see also Meyer, TT, 9/13/2011 a.m., at 105-106. Accordingly, since 2008, HRB’s average DDIY sales price has declined, while the average revenue per paid customer has remained roughly the same. 33 In the defendants’ submissions to the Antitrust Division of the DOJ prior to this litigation, the defendants appeared to emphasize this “value” and “premium” distinction as the basis for their definition of the relevant market. See GX 135 at 14-15; GX 629 at 18-30. As a result, the government accuses the defendants of having “tacked back and forth” regarding their proposed relevant market definition. Pl.’s Post-Trial Mem. at 1-2. While the Court agrees that the import of the hearing testimony about value and premium products was not always clear, the defendants’ counsel clarified during closing arguments that the “only real relevance” of the premium versus value distinction was to show that HRB and TaxACT are not closest the competitors for the purposes of unilateral effects analysis. Defs.’ Closing Argument, TT, 10/3/2011 a.m., at 93-94. 70 See GX 296-7 (“Digital Tax Solutions FY11 Actual Deep Dive”) at 1; Meyer, TT, 9/13/11 a.m., at 107-108. Further, the evidence discussed above indicating direct price and feature competition between HRB and TaxACT negates the conclusion that they operate in separate value and premium segments of the market. There are certainly occasional references to different pricing levels in the defendants’ documents. See GX 20 at 11 (HRB document noting {redacted}) (emphasis added). This hardly means that the companies are not in close competition, however. Rather, as Mr. Dunn’s testimony reflects, TaxACT competes with capital-rich HRB and Intuit by offering high-quality products at substantially lower prices. See Dunn, TT, 9/7/11 p.m., at 71-72 (noting that rather than attempting to outspend its richer competitors on marketing, TaxACT’s growth strategy has depended on providing “great customer service, a great product, and a great customer experience” for a much lower price, including free). Id. This type of healthy competition benefits taxpaying consumers. The fact that Intuit may be the closest competitor for both HRB and TaxACT also does not necessarily prevent a finding of unilateral effects for this merger. See Areeda & Hovenkamp, ¶ 914, 77-80 (explaining that the merging parties need not be the closest rivals for there to be unilateral anticompetitive effects); see also Commentary on the Horizontal Merger Guidelines (2006) at 28 (“A merger may produce significant unilateral effects even though a non-merging product is the ‘closest’ substitute for every merging product . . .”). Using a simple estimate of diversion based on market share would indeed suggest that HRB and TaxACT are each other’s second closest rivals after Intuit.34 See GX 121 (Warren-Boulton Rep.) at 44 (explaining that 34 The relevance of the diversion estimates provided by the expert economists to the unilateral effects analysis is discussed more fully below. 71 using market share to estimate diversion is a “benchmark” assumption in standard empirical models of consumer demand). iv. Merged Company’s Combined Market Share Another argument that the defendants present against a likelihood of unilateral effects is that, in their view, unilateral effects cannot be demonstrated where the combined firm’s market share does not surpass a certain threshold. The defendants point out that in Oracle, the court stated that “[a] presumption of anticompetitive effects from a combined share of 35% in a differentiated products market is unwarranted. Indeed, the opposite is likely true.” 331 F. Supp. 2d at 1123. The Oracle court stated that “[t]o prevail on a differentiated products unilateral effects claim, a plaintiff must prove a relevant market in which the merging parties would have essentially a monopoly or dominant position.” Id. Some commentators have criticized this standard, however, because “impermissible price increases . . . can be achieved on far lower market shares” than Oracle’s standard evidently requires. Areeda & Hovenkamp ¶ 914, at 84. Indeed, Judge Brown’s subsequent opinion from this Circuit in Whole Foods implied that a market definition itself may not even be required for proving a Section 7 violation based on unilateral effects. See Whole Foods, 548 F.3d at 1036. In a footnote, Judge Brown explained that “a merger between two close competitors can sometimes raise antitrust concerns due to unilateral effects in highly differentiated markets. In such a situation, it might not be necessary to understand the market definition to conclude a preliminary injunction should issue.”35 Id. at 35 “As a matter of applied economics, evaluation of unilateral effects does not require a market definition in the traditional sense at all.” Areeda & Hovenkamp ¶ 913a, at 66. This is so because unilateral effects analysis focuses on measuring a firm’s market power directly by “estimating the change in residual demand facing the post-merger firm. ‘Residual demand’ refers to the demand for a firm’s goods after the output of all other competing firms has been taken into account.” Id. at 63. If market power itself can be directly measured or estimated reliably, then in theory market definition is superfluous, at least as a matter of economics, because “[i]dentifying a market and computing market shares provide an indirect means for measuring market power.” Id. ¶ 532a at 242-43; see also id. ¶ 521c. The 2010 revisions to the Merger Guidelines also appear to reflect this understanding. See Merger Guidelines § 4 (“The Agencies’ analysis need not start with market definition. Some of the analytical tools used by 72 n.1 (citation omitted). The Court therefore declines the defendants’ invitation, in reliance on Oracle, to impose a market share threshold for proving a unilateral effects claim.36 v. Post-Merger Dual Brand Strategy HRB’s plans for the post-merger company raise anticompetitive questions. Post-merger, HRB’s stated plan is to maintain both the HRB and TaxACT brands –with the HRB-brand focusing on higher priced-products and the TaxACT brand focusing on the lower-priced products. See Bennett, TT, 9/6/11 a.m., 101-102; DX 1005 at 1. HRB’s general pre-merger pricing strategy has been to price its products a bit below Intuit’s products. Bennett, TT, 9/6/11 a.m., at 99. Part of HRB’s post-merger strategy, however, appears to involve raising prices on HRB-branded products. Under this two-brand strategy, HRB would price its “premium” HRBbranded products equal to or above Intuit’s prices. See Bennett, TT, 9/6/11 a.m., 101-102; DX 1005 at 1. At the same time, the company would “offer TaxACT as its free and value brand.” DX 17 (Meyer Rep.) at 78. Yet, the defendants have never convincingly explained how this two-brand strategy would work in practice because defendants have repeatedly emphasized how the Agencies to assess competitive effects do not rely on market definition, although evaluation of competitive alternatives available to customers is always necessary at some point in the analysis.”). As a legal matter, however, a market definition may be required by Section 7 of the Clayton Act. See Brown Shoe, 370 U.S. at 324 (“[D]etermination of the relevant market is a necessary predicate to a finding of a violation of the Clayton Act because the threatened monopoly must be one which will substantially lessen competition ‘within the area of effective competition.’ Substantiality can be determined only in terms of the market affected. The ‘area of effective competition’ must be determined by reference to a product market (the ‘line of commerce’) and a geographic market (the ‘section of the country’).” ) (internal citation omitted); see also Heinz, 246 F.3d at 719 n.17 (“Courts interpret ‘line of commerce’ [in the language of the Clayton Act] as synonymous with the relevant product market.”). The Court is not aware of any modern Section 7 case in which the court dispensed with the requirement to define a relevant product market, although Judge Brown’s opinion in Whole Foods may be read to endorse this possibility in accordance with the evolving understandings in economics. See Whole Foods, 548 F.3d at 1036 (Brown, J.) (stating that the Baker Hughes analytical framework, which “rests on defining a market and showing undue concentration in that market,” “does not exhaust the possible ways to prove a § 7 violation on the merits”). 36 The Commentary on the Merger Guidelines, for its part, explains that while “[a]s an empirical matter, the unilateral effects challenges made by the Agencies nearly always have involved combined shares greater than 35%,” “the Agencies may challenge mergers when the combined share falls below 35% if the analysis of the mergers’ particular unilateral competitive effects indicates that they would be likely substantially to lessen competition.” Commentary on the Horizontal Merger Guidelines (2006) at 26. “Combined shares less than 35% may be sufficiently high to produce a substantial unilateral anticompetitive effect if the products are differentiated and the merging products are especially close substitutes . . . .” Id. 73 important “free” product offerings are for all DDIY brands. See DFF ¶ 185 (“Free is a highly profitable method of acquiring customers for H&R Block.”); DX 600 at 10 (HRB Board of Directors presentation for merger approval stating that after the merger TaxACT would be the “low cost value provider focused on free” but that the company would “[c]ontinue to offer a free product in the HRB brand to drive client acquisition”). Part of the government’s concern with HRB’s two-brand strategy is that the incentives for the combined firm in marketing and developing the TaxACT product would be quite different from the incentives that exist in the current market. HRB may feel comfortable raising its “premium” prices because it knows that consumers looking for lower-cost DDIY options would be most likely to migrate to TaxACT, the established “value leader” in the market. Since HRB will also control TaxACT post-merger, however, HRB can still ensure that TaxACT’s value proposition does not get “too good” and undermine the paid HRB products with the highest profit margins. For example, HRB might restrict the features of TaxACT’s free and low-cost products to ensure they do not cannibalize sales of HRB’s higher priced offerings. Indeed, assuming that there are high barriers to entry and expansion, this strategy would appear logical because it would maximize HRB’s profit per customer. Post-merger, TaxACT will not have the same incentives it has today to develop robust free and low-cost offerings that can compete with the functionality offered by HRB and Intuit. See Warren-Boulton, TT, 9/8/11 p.m., at 32-33. Thus, this merger could potentially have the effect of stifling price and feature competition compared with maintaining TaxACT as an independent firm. vi. Merger Simulation Shows Likely Unilateral Price Increase The government’s expert economist, Dr. Warren-Boulton, did a merger simulation analysis that suggests a unilateral price increase is likely. Warren-Boulton, TT, 9/9/11 a.m., at 5- 74 11; GX 121 (Warren-Boulton Rep.) at 52. The key factors in this simulation are HRB and TaxACT’s price-cost margins and the diversion ratios between their products. Cf. Swedish Match, 131 F. Supp. 2d at 169 (“High margins and high diversion ratios support large price increases, a tenet endorsed by most economists.”). (a). Diversion Ratios Between the Merging Parties’ DDIY Products As explained above, the diversion rate from TaxACT to HRB measures the proportion of customers that would leave TaxACT in response to a price increase and switch to HRB. Dr. Warren-Boulton’s report explains that higher diversion rates between merging parties “allow the firms to recapture more lost sales following a price increase, and therefore lead to greater upward pricing pressure and post-merger unilateral price increases.” GX 121 (Warren-Boulton Rep.) at 44. Dr. Warren-Boulton estimated diversion ratios from two sources: the parties’ DDIY market share data and the IRS switching data.37 Id. at 44-48. By assuming diversion rates in accordance with market share, Dr. Warren-Boulton estimated the diversion rate from TaxACT to HRB to be 12 percent and from HRB to TaxACT to be 14 percent. Id. at 44-45. Dr. Warren-Boulton notes that these diversion estimates likely underestimate what the actual post-merger diversion rates will be since the merged company will likely implement marketing strategies to keep customers within the umbrella of the combined company. Id. at 45. Dr. Warren-Boulton estimated diversion ratios using IRS switching data as well. As discussed above in Section III.A.3.a, he also used this switching data to test the relevant market definition. As previously noted in that prior discussion, switching data is not equivalent to diversion, since diversion measures switching in response to a price increase as opposed to all 37 Dr. Warren-Boulton declined to rely on the defendants’ proposed diversion data, derived from their consumer surveys, for the reasons already discussed supra in Section III.A.3. 75 switching generally. In particular, Dr. Warren-Boulton found that switching data is especially likely to overstate diversion from DDIY products to assisted preparation. Id. at 46-47. Therefore, Dr. Warren-Boulton discounted the switching rates from DDIY to assisted by half to correct for this effect.38 Id. After this correction, Dr. Warren-Boulton calculated estimated diversion rates from TaxACT to HRB and from HRB to TaxACT of 12 percent. Id.at 47-48. (b). Price-Cost Margins The next step in his analysis was to estimate the firms’ price-cost margins. “All else equal, higher margins lead to greater unilateral price increases because the value of recaptured sales is higher.” Id. at 48. Using a procedure described in his report, Dr. Warren-Boulton estimated {that the merging parties have high margins}. Id. at 49. The merger simulation also required quantities of units sold and average revenue per unit. Dr. Warren-Boulton obtained this data from the companies’ submissions. Id. at 50. (c). Simulation Results Using all of these data, Dr. Warren-Boulton performed a linear demand Bertrand model simulation. Id. at 51. Unless there are significant efficiencies from the merger that are passed on to consumers, this simulation predicts a unilateral price increase.39 Id. at 52. Assuming diversion ratios according to market share, the model predicts TaxACT’s price will increase by 12.2 percent and HRB’s price by 2.5 percent. Id. Assuming diversion ratios based on the IRS switching data as discussed above, the model predicts TaxACT’s price will increase by 10.5 percent and HRB’s price by 2.2 percent. Id. 38 As a basis for this conclusion that switching data overstates diversion and for his choice to discount the DDIY-toassisted switching rate by half, Dr. Warren-Boulton relies upon HRB documents that suggest that more than half of switching from DDIY to assisted occurs for reasons unrelated to price, such as a change in tax complexity. GX 121 (Warren-Boulton Rep.) at 46 n.128 (citing GX 635, GX 126). He also relies on IRS data showing that customers switching from DDIY to assisted were twice as likely to have a complexity increase as taxpayers who stayed within DDIY. Id. at 47. 39 As discussed in Section III.B.2.d below, the Court finds most of defendants’ claimed efficiencies are not mergerspecific or unverifiable. 76 (d). Critique of the Simulation’s Unilateral Effects Results The defendants attack Dr. Warren-Boulton’s simulation on several grounds. The defendants reiterate their critique that switching data is an inappropriate proxy for diversion data. Further, defendants criticize the way in which Dr. Warren-Boulton discounted the switching rates from DDIY products to assisted preparation. See Warren-Boulton, TT, 9/9/9 p.m., at 6065. In addition, the defendants contend that Dr. Warren-Boulton’s simulator model is flawed because it will always predict a price increase with any positive diversion and because the model is “static,” does not take various factors into account, such as the parties’ different products, innovation, and marketing, and would never predict that a firm would offer free products, even though free products are a staple of the industry. DX 17 (Meyer Rep.) at 74-75. The Court agrees that Dr. Warren-Boulton’s discounting by half of the switching data from DDIY to assisted appears imprecise. Dr. Warren-Boulton clarified in his report, however, that “the model still predicts significant unilateral harm when non-discounted switching rates are used to approximate diversion rates.” GX 121 (Warren-Boulton Rep.) at 47. Further, and more importantly, Dr. Warren-Boulton also estimated diversion ratios based on market share and the Court has concluded above that DDIY is the appropriate relevant product market.40 As for the defendants’ critiques about Dr. Warren-Boulton’s economic model itself, Dr. Warren-Boulton addressed these directly. First, insofar as the model will predict at least some price increase absent efficiencies with any positive diversion ratios, Dr. Warren-Boulton explained that outcome is fully consistent with correct economic theory. GX 665 (Warren- 40 The defendants suggest that Dr. Warren-Boulton’s reliance on market share as an estimate of diversion ratios is somewhat circular in that his market shares derive from his market definition, which, in turn, relied on his use of switching data as a proxy for diversion ratios. DX 17 (Meyer Rep.) at 76. As discussed above, however, the Court’s finding that DDIY is the correct relevant product market is not dependent on Dr. Warren-Boulton’s analysis. 77 Boulton Reply Rep.) at 14 (“Economic theory concludes that absent merger specific efficiencies, a merger between competing firms will cause the merging firms to increase their prices by at least some amount. Thus, it is not a deficiency, but a strength, of merger simulation models that they reflect this aspect of economic reality.”). In response to the critique that his “static” model would never predict that companies would offer free products, Dr. Warren-Boulton contends that because free DDIY products are often packaged with other paid products, these “free” products actually provide the companies with a positive average revenue per free unit, which his model does take into account. See id. at 14-15. As for the remaining critiques that the model does not factor in marketing or innovation, Dr. Warren-Boulton replies that any model is inherently a simplification of the real world, but there is no reason to assume these factors negate the price effect findings of the model. Id. The Court finds that the merger simulation model used by the government’s expert is an imprecise tool, but nonetheless has some probative value in predicting the likelihood of a potential price increase after the merger. The results of the merger simulation tend to confirm the Court’s conclusions based upon the documents, testimony, and other evidence in this case that HRB and TaxACT are head-to-head competitors, that TaxACT’s competition has constrained HRB’s pricing, and that, post-merger, overall prices in the DDIY products of the merged firms are likely to increase to the detriment of the American taxpayer. vii. Repositioning Unlikely to Defeat Unilateral Price Increase Repositioning by smaller competitors in response to a unilateral price increase is unlikely for the same reasons discussed above regarding barriers to entry and expansion. See Merger Guidelines § 6.1 (“Repositioning is a supply-side response that is evaluated much like entry, with consideration given to timeliness, likelihood, and sufficiency.”). 78 Repositioning by Intuit is also unlikely due to the coordinated effects incentives discussed above. The Merger Guidelines make clear that a unilateral price increase may be defeated where “non-merging firms [are] able to reposition their products to offer close substitutes for the products offered by the merging firms.” Merger Guidelines § 6.1. Since the Court has already found that HRB and Intuit would have coordinated pricing incentives postmerger, that finding implies that repositioning by Intuit would not prevent HRB from raising prices. By relying on its finding of coordinated effects to predict the likelihood of repositioning by Intuit, the Court acknowledges that its unilateral effects finding is not strictly “unilateral” in the sense that it does take coordination into account. The case law and the Merger Guidelines, however, require that “repositioning” be considered in assessing unilateral effects, and the repositioning inquiry necessarily entails a consideration of the likely actions of other competitors in response to a price increase. See CCC Holdings, Inc., 605 F. Supp. 2d at 67 (noting that the distinction between coordinated and unilateral effects “has more significance in law than it does in economics” and citing expert testimony describing the distinction as “artificial”). viii. Finding Unilateral Anticompetitive Effects Likely On balance, and considering the evidence as a whole, the Court finds that, absent efficiencies, the plaintiff has demonstrated a reasonable likelihood of unilateral effects by a preponderance of the evidence. See Swedish Match, 131 F. Supp. 2d at 169 (finding likelihood of unilateral price increase where merger would eliminate one of the larger merging firm’s “primary direct competitors,” “the third largest selling” brand “that has consistently played a role in constraining the price” of the larger firm’s products); see also Staples 970 F. Supp. at 1083 (finding anticompetitive effects where the “merger would eliminate significant head-to-head competition between the two lowest cost and lowest priced firms in the . . . market.”). 79 The Court will now turn to the defendants’ final rebuttal argument – the existence of significant, merger-specific efficiencies. d. Post-Merger Efficiencies One of the key benefits of a merger to the economy is its potential to generate efficiencies. See Heinz, 246 F.3d at 720. As the Merger Guidelines recognize, merger-generated efficiencies can “enhance the merged firm’s ability and incentive to compete, which may result in lower prices, improved quality, enhanced service, or new products.” Merger Guidelines § 10. Courts have recognized that a showing of sufficient efficiencies may rebut the government’s showing of likely anticompetitive effects. Heinz, 246 F.3d at 720. High market concentration levels require “proof of extraordinary efficiencies,” however, and courts “generally have found inadequate proof of efficiencies to sustain a rebuttal of the government’s case.” Id. (citation omitted). “[T]he court must undertake a rigorous analysis of the kinds of efficiencies being urged by the parties in order to ensure that those ‘efficiencies’ represent more than mere speculation and promises about post-merger behavior.” Id. at 721. As the Merger Guidelines explain, “[c]ognizable efficiencies are merger-specific efficiencies that have been verified and do not arise from anticompetitive reductions in output or service.” Merger Guidelines § 10. Efficiencies are inherently “difficult to verify and quantify” and “it is incumbent upon the merging firms to substantiate efficiency claims” so that it is possible to “verify by reasonable means the likelihood and magnitude of each asserted efficiency, how and when each would be achieved (and any costs of doing so), how each would enhance the merged firm’s ability and incentive to compete, and why each would be merger-specific.” Id. In other words, a “cognizable” efficiency claim must represent a type of cost saving that could not be achieved 80 without the merger and the estimate of the predicted saving must be reasonably verifiable by an independent party. The defendants claim that “H&R Block’s primary motivation for the TaxACT acquisition is to achieve significant synergies that will enable H&R Block to provide better products at a lower price and to compete more effectively.”41 Defs.’ Post-Trial Mem. at 24. The defendants predict that they will achieve over ${redacted} million in annual efficiencies in ten different areas.42 Id. at 24-25. The chart below summarizes the defendants’ claimed efficiencies and predicted annual cost savings: 41 “Cognizable efficiencies” are a subset of “synergies.” “Synergies” refer more generally to any business performance benefits that result from the merger of two companies. See Zmijewski, TT, 9/19/11 a.m., at 99. 42 Originally, the defendants claimed 11 efficiencies, including an efficiency related to {redacted}. This task is “really not an efficiency” but “an additional cost,” Dunn, TT, 9/8/11 p.m. (sealed) at 7, and defendants do not reference it in their proposed findings of fact. DFF ¶ 291. 81 Efficiency Description 1. Online IT {redacted} 2. Emerald Card Allowing TaxACT’s prepaid debit card offerings to be fulfilled through HRB’s bank Funding TaxACT’s refund anticipation checks through HRB’s bank ${redacted} million {redacted} ${redacted}million 5. {redacted} {redacted} ${redacted} million 6. {redacted} {redacted} ${redacted} million 7. Corporate Website 8. Software IT {redacted} ${redacted} million {redacted} ${redacted} million 9. Download Fulfillment 10. {redacted} {redacted} ${redacted} million {redacted} ${redacted}million 3. H&R Block Bank Refund Anticipation Checks 4. {redacted} Estimated Annual Cost Saving ${redacted} million ${redacted} million DFF ¶ 292; see also DX236-007. Dr. Mark E. Zmijewski, an expert witness for the government, analyzed the defendants’ alleged efficiencies and concluded that – with the exception of {one efficiency related to eliminating third-party contracts} – the proposed efficiencies identified by the defendants are either not merger-specific or not verifiable.43 See generally GX 664 (Zmijewski Rep.). The Court agrees with Dr. Zmijewski that the defendants have not demonstrated that their claimed efficiencies are merger-specific. If a company could achieve certain cost savings 43 Dr. Zmijewski is a professor of accounting and deputy dean at The University of Chicago Booth School of Business and a founder and principal of Navigant Economics, a consulting firm. GX 664 (Zmijewski Rep.) at 5. He holds a Ph.D. in accounting. Id. 82 without any merger at all, then those stand-alone cost savings cannot be credited as mergerspecific efficiencies. The defendants must show that their “efficiencies . . . cannot be achieved by either company alone because, if they can, the merger’s asserted benefits can be achieved without the concomitant loss of a competitor.” Heinz, 246 F.3d at 722. For example, if HRB’s {redacted} are not running in the most efficient, cost-effective manner, it is hard to see why a merger with TaxACT is necessary to improve their cost structure. The reasons HRB claims it has higher {redacted} costs than TaxACT include (1) that TaxACT has lower labor costs in Cedar Rapids than HRB has in Kansas City and (2) that TaxACT is simply more cost conscious. Bowen, TT, 9/15/11 p.m., (sealed), at 104-105. Plainly, then, HRB could therefore achieve at least some of the {redacted} cost savings on its own – by relocating {redacted} and taking a more cost conscious attitude toward them. Likewise, the efficiencies related to bringing HRB’s outsourced {redacted} functions in-house are unlikely to be wholly merger-specific. Similarly, the defendants’ IT-related efficiencies, which account for the largest efficiency claims, are not entirely merger-specific either. Both TaxACT and HRB witnesses testified that {redacted} – suggesting that the platform consolidation would result in at least some mergerspecific efficiencies. See Dunn, TT, 9/8/11 p.m. (sealed), at 16-17; Bowen, TT, 9/15/11 p.m. (sealed), at 67-68. One way in which {redacted}. Dunn, TT, 9/8/11 p.m. (sealed), at 16-17; Bowen, TT, 9/15/11 p.m. (sealed), at 67-68; Bowen, TT, 9/19/11 a.m., at 12. Thus, the IT consolidation efficiency actually can be thought of as entailing two distinct consolidations: (1) {redacted} and (2) HRB’s platform will be merged with TaxACT’s platform. Bowen, TT, 9/19/11 a.m., at 12. Yet the claimed IT efficiency is not discounted for whatever savings HRB could obtain by {performing the first consolidation} on its own – an option the company considered in the past but did not adopt – and the defendants did not present evidence explaining 83 why, as a technical matter, {performing the first consolidation} would not be feasible or, in fact, would not be more feasible than {the double consolidation}. Bowen, TT, 9/19/11 a.m., at 12; 9/15/11 p.m. (sealed) at 75. The IT efficiencies also apparently account for cost reductions associated with TaxACT’s more cost-conscious culture and practices. See Dunn, TT, 9/8/11 a.m. (sealed), at 5 (“for Block to achieve these [efficiencies] would require them to come up with an entirely different corporate culture {redacted}.”). Even if the efficiencies were entirely merger-specific, many of them are also not independently verifiable. As Dr. Zmijewski explained, for the various efficiencies that involve the activities now performed by HRB or its vendors that are proposed to be transferred to TaxACT, TaxACT’s predicted cost figures for taking over these activities were not based on an analysis of facts that could be verified by a third party. Instead, TaxACT based its cost estimates on management judgments. GX 664 (Zmijewski Rep.) at 22-25. By comparison, HRB’s estimated costs for the relevant activities were rooted in accounting and planning documents prepared in the ordinary course of business. The testimony at the hearing confirmed that TaxACT’s recurring cost estimates were largely premised on its managers experiential judgment about likely costs, rather than a detailed analysis of historical accounting data. See, e.g., Dunn, TT, 9/8/11 p.m. (sealed), at 28-31. While reliance on the estimation and judgment of experienced executives about costs may be perfectly sensible as a business matter, the lack of a verifiable method of factual analysis resulting in the cost estimates renders them not cognizable by the Court. If this were not so, then the efficiencies defense might well swallow the whole of Section 7 of the Clayton Act because management would be able to present large efficiencies based on its own judgment and the Court would be hard pressed to find otherwise. The difficulty in substantiating efficiency claims in a verifiable 84 way is one reason why courts “generally have found inadequate proof of efficiencies to sustain a rebuttal of the government’s case.” Heinz, 246 F.3d at 720 (citation omitted); see also Staples, 970 F. Supp. at 1089 (finding “defendants failed to produce the necessary documentation for verification” of efficiencies). Particular scrutiny of HRB’s efficiencies claims is also warranted in light of HRB’s historical acquisitions. In 2006, HRB acquired a software company called TaxWorks, which was renamed “RedGear.” Bowen, TT, 9/15/11 p.m. (sealed), at 84. For the RedGear acquisition, which was much smaller in scale than the proposed TaxACT deal, HRB projected a total of ${redacted} million in efficiencies over three years. GX 1459 (February 2009 “Taxworks Financial Analysis”) at 5. HRB failed to achieve these {efficiencies} {redacted}. Id. In this case, the efficiency estimates are much more aggressive, in that defendants are claiming approximately ${redacted} million in efficiencies for 2013 and ${redacted} million in annual savings going forward thereafter, as opposed to ${redacted} million over three years. See Bowen, TT, 9/15/11 p.m. (sealed), at 77-78. While HRB has attempted to learn from the mistakes of the RedGear acquisition, id. at 85-87, the Court finds that this history only underscores the need for any claimed efficiencies to be independently verifiable in order to constitute evidence that can rebut the government’s presumption of anticompetitive effects. Considering all of the evidence regarding efficiencies, the Court finds that most of the defendants’ claimed efficiencies are not cognizable because the defendants have not demonstrated that they are merger-specific and verifiable.44 44 In addition, the defendants have not addressed how much of the claimed efficiencies would be passed through to consumers. See Staples, 970 F. Supp. at 1090 (analyzing projected pass-through rate for claimed efficiencies). 85 IV. CONCLUSION The Court concludes that the proposed merger between HRB and TaxACT violates Section 7 of the Clayton Act because it is reasonably likely to cause anticompetitive effects. The law of this Circuit supports this conclusion. In Heinz, the Court of Appeals reversed a district court’s denial of a preliminary injunction against a merger involving the second- and thirdlargest jarred baby food companies. 246 F.3d at 711-12. After noting the high barriers to entry and high HHI figures that characterized the market, the D.C. Circuit observed that “[a]s far as we can determine, no court has ever approved a merger to duopoly under similar circumstances.” Id. at 717. The situation in this case is similar. The government established a prima facie case indicating that anticompetitive effects are likely to result from the merger. The defendants have not made a showing of evidence that rebuts the presumption of anticompetitive effects by demonstrating that the government’s market share statistics give an inaccurate account of the merger’s probable effects on competition in the relevant market. To the contrary, the totality of the evidence confirms that anticompetitive effects are a likely result of the merger, which would give H&R Block and Intuit control over 90 percent of the market for digital do-it-yourself tax preparation products. Accordingly, the Court will enjoin H&R Block’s proposed acquisition of TaxACT. An appropriate Order will accompany this Memorandum Opinion. Beryl A. Howell DATED: November 10, 2011 /s/ BERYL A. HOWELL United States District Judge 86

Some case metadata and case summaries were written with the help of AI, which can produce inaccuracies. You should read the full case before relying on it for legal research purposes.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.