Monday, January 27, 2014
Posted by William Kolasky
Anyone who has been through a merger investigation at either the Justice Department or the Federal Trade Commission knows that both agencies rely heavily on the testimony of customers both in their initial decision as to whether to issue a Second Request and in their final decision on whether or not to challenge a merger. The FTC’s most recent compilation of Horizontal Merger Investigation Data, covering the fiscal years 1996 to 2011, for example, reports that the FTC challenged almost all of the mergers it reviewed where it had received strong customer complaints (111 out of 114), but fewer than half the mergers as to which there were no strong customer complaints (53 out of 122).
It is always interesting, therefore, to see how little weight judges generally give the testimony of customers when a merger challenge gets to court. Three examples come to mind. In the Justice Department’s challenge to the Sungard/Comdisco merger involving shared disaster recovery systems for mainframe computers in 2001, Judge Ellen Segal Huvelle gave little credence to the 50 declarations of customers the government proffered to show that customers would not switch to internal recovery systems in response to a SSNIP. She noted that defendants had proffered an even larger number of customers declarations (90) saying just the opposite and that the 140 declarations the two sides had presented represented fewer than 5% of all customers, with no assurance that these 5% were representative. Similarly, in the Justice Department’s 2004 challenge to Oracle’s proposed acquisition of its competitor, PeopleSoft, Judge Vaughn Walker wrote that he found the testimony of the ten customers proffered to support its narrow market definition “largely unhelpful.”  While acknowledging that the ten witnesses were plainly very sophisticated buyers of enterprise software, he observed that none of them had done a careful cost-benefit analysis of the kind he would have expected a sophisticated buyer to do in order to support their “unsubstantiated apprehensions” that the proposed merger would enable Oracle to raise prices on the software products as to which it competed with PeopleSoft. Finally, in the Justice Department’s more recent 2011 challenge to H & R Block’s attempted acquisition of one of its two leading competitors for tax preparation software, TaxACT, Judge Beryl Howell rejected the defendants’ effort to rely on a survey of over 1,000 customers to show that they would switch to other tax preparation methods in the event of a SSNIP, finding serious methodological problems with the proffered survey.
Judge William Orrick’s opinion in Bazaarvoice is the latest example of a judge finding unpersuasive customer testimony as to the likely effect of a merger on competition. In challenging the merger, the government made out its prima facie case by defining a very narrow product market—Ratings and Review platforms for online commerce—in which the evidence showed that the two merging parties, Bazaarvoice and PowerReview, had a combined share of more than 60%, or nearly 100% if you exclude in-house platforms, which the government argued were not a viable alternative for many customers. The government buttressed its prima facie case with internal documents showing that each company viewed the other as its closest competitor and that they both anticipated that the merger would provide “relief from . . . price erosion” and permit “margin expansion.”
In its effort to rebut the government’s prima facie case, Bazaarvoice relied heavily on customer testimony, arguing that “none of the more than 100 current, former and potential customers who testified in the case believed that the acquisition had or would harm them.” Judge Orrick was unmoved. While acknowledging that “the customers were the most credible sources of information on their need for, use of and substitutability of social commerce products”—and, indeed, relying on that testimony in accepting the government’s narrow market definition--he found that their testimony on the impact and likely effect of the merger “was speculative at best” and was therefore “entitled to virtually no weight.” As he explained, because “products such as R&R platforms are relatively inexpensive in comparison to a company’s operating budget and have relatively long contracts . . . [c]onsidering competitive alternatives for R&R is not part of customer’s day-to-day activities.” Most of the customers testified, therefore, that “they had never given any thought to the merger” and did “not have much current information about” its likely effects. The judge dismissed the opinions of those who had thought about the merger on the ground that they had “not been privy to most of the evidence presented to the court, including that of the economic experts, and therefore had only a “narrow perspective of the economic and legal questions before the court,” and “an idiosyncratic understanding” of the R&R market.
In considering the implications of Judge Orrick’s rejection of Bazaarvoice’s effort to rely on customer testimony to rebut the government’s case, it is easy to identify reasons why his ruling may have limited impact. First, the government’s prima facie case was very strong—based not just on market shares, but also on detailed win-loss data and highly damaging statements in the merging parties’ internal documents. As the D.C. Circuit held in Baker Hughes, in language Judge Orrick quoted: “The more compelling the prima facie case, the more evidence the defendant must present to rebut it successfully.” Second, in cases where the products or services at issue are more central to their customers’ business and are purchased more regularly, a court would have a much harder time finding that the customers had not given any thought to a merger potentially affecting them or had insufficient information to inform their views as to its likely effect on their business. Third, as in the earlier cases cited above, the customers who testified apparently did not back up their opinion testimony with the type of detailed analysis that would have been required for Judge Orrick to find their testimony persuasive.
That said, Judge Orrick’s opinion should teach all of us who practice antitrust law—whether at the agencies or for parties appearing before them—the importance assuring that a customer’s opinion testimony is supported by both facts and analysis. His opinion also shows the importance—both at the agency level and in court--of testing the customers’ opinions against the objective evidence. For example, a statement by a customer that it views the merging parties as “next best substitutes” should be greeted skeptically if the data show that it buys from most of its requirements from one of the two merging parties and the rest from another competing supplier, rather than from the other merging party.
 Federal Trade Commission, Horizontal Merger Investigation Data: Fiscal Years 1996 – 2011, Tables 7.1 & 7.2 (January 2013).
 United States v. Sungard Data Systems, Inc., 172 F. Supp.2d 172 (D.D.C.2001)
 United States v. Oracle Corp., 331 F. Supp. 2d 1098 (N.D. Cal. 2004).
 United States v. Bazaarvoice, Inc., Case No. 13-cv-00133-WHO (N.D. Cal. Jan. 8, 2014).
 United States. v. H & R Block, Inc., 833 F. Supp. 2d 36 (D.D.C. 2011)
 Slip Op. at 29.
 Id. at 116.
 Id. at 138.
 Id. at 131, quoting United States v. Baker Hughes Inc., 908 F.2d 981, 991 (D.C. Cir. 1989).