Opinion
MDL 3010
08-12-2021
IN RE: DIGITAL ADVERTISING ANTITRUST LITIGATION
TRANSFER ORDER
Karen K. Caldwell, Chair
Before the Panel:
Defendants Google LLC, Alphabet Inc., and YouTube, LLC (together, "Google") move under 28 U.S.C. § 1407 to centralize this antitrust litigation in the Northern District of California. The litigation currently consists of 19 actions pending in 16 districts, as listed on Schedule AThe actions concern Google's alleged monopolization and suppression of competition in online display advertising - essentially, the marketplace for the placement of digital display ads on websites and mobile apps, The parties describe the principal participants in online display advertising as advertisers seeking to place ads on the internet, online content providers such as news sites offering ad space alongside digital content, and high-speed electronic trading venues called "exchanges" that advertisers and online publishers use to manage the buying and selling of ad space. The actions allege that Google runs the largest ad exchange and has engaged in unlawful acts to suppress competition, causing injuries to advertisers and publishers that participate in its exchange by imposing supracompetitive pricing and depriving them of revenue. Plaintiffs in all actions seek declaratory and equitable relief under federal or state antitrust laws to stop the alleged conduct and damages. The parties refer to the actions by the type of plaintiff involved - namely, the advertiser actions (three actions), the publisher actions (sixteen actions), and the state attorneys general action filed by 15 states in the Eastern District of Texas CState of Texas" or "State Action").
Google's motion lists 20 actions for centralization. After the motion was filed, one action (Organic Panaceas) was closed following consolidation for all purposes with another action in the Northern District of California. Since the filing of Google's motion, the Panel also has been notified of one potential tag-along action.
The 15 state plaintiffs are Texas, Alabama, Arkansas, Florida, Idaho, Indiana, Kentucky, Mississippi, Missouri, Montana, Nevada, North Dakota, Puerto Rico, South Dakota, and Utah. Two other states - Louisiana and South Carolina - have moved to intervene as additional plaintiffs. The state plaintiffs recently submitted a supplemental brief stating that they have filed a motion for leave to file a second amended complaint in the underlying court that, if granted, would eliminate their request for damages under federal antitrust law, along with adding the putative intervenors as plaintiffs. The proposed amended complaint makes the same factual allegations and asserts the same federal antitrust violations under the Sherman Act as the operative first amended complaint. Google filed a response to the supplemental information stating that Google has reserved its right to oppose the amendment and arguing that centralization remains appropriate. We have considered these post-hearing submissions, along with two supplemental briefs submitted by plaintiffs in the private actions, in deciding the motion for centralization.
I.
Defendant Facebook supports centralization of all actions in the Northern District of California or, alternatively, the Southern District of New York. On the plaintiffs' side, there are varying positions on the threshold issue of whether centralization of these actions is warranted and the appropriate transferee district. As to plaintiffs in the advertiser actions, plaintiff in one Northern District of California action (SPX Total Body Fitness) supports centralization of all actions in the Northern District of California. Plaintiffs in the Northern District of California In re Google Digital Advertising Antitrust Litigation action oppose centralization. And plaintiffs in the District of Columbia Cliffy Care action oppose inclusion of their action in any MDL. As to plaintiffs in the publisher actions, all request that the State of Texas action be excluded from any MDL to ensure its expeditious resolution. Beyond that, however, their positions vary significantly. Plaintiffs in the Northern District of California In re Google Digital Publisher Antitrust Litigation oppose centralization and, alternatively, request their district as the transferee forum. Plaintiffs in 13 regional newspaper actions oppose centralization of their actions and, alternatively, request the Eastern District of Texas, the Southern District of New York, or the District of the District of Columbia. Plaintiffs in the Southern District of New York Associated Newspapers action stated at oral argument that they oppose centralization and, in the alternative, assert that the actions should proceed in three separate districts - specifically, the Southern District of New York for the 15 publisher actions; the Northern District of California for the three advertiser actions; and the Eastern District of Texas, where the State of Texas action would proceed in its original forum independently from any MDL. The state plaintiffs oppose inclusion of State of Texas in any MDL and take no position on centralization of the other actions; alternatively, they request that all actions be centralized in the Eastern District of Texas. Lastly, two non-party trade organizations representing news entities and other online content providers - News Media Alliance and Digital Content Next ("DCN") - filed interested party briefs supporting a separate publisher MDL in the Southern District of New York and requesting exclusion of State of Texas from any MDL. DCN also suggested the Eastern District of Texas in its oral argument notice.
In the Panel briefing and their oral argument notice, plaintiffs in Associated Newspapers supported centralization in the first instance, proposing the three groupings described above.
On the basis of the papers filed and the hearing session held, we find that these actions involve common questions of fact, and that centralization in the Southern District of New York will serve the convenience of the parties and witnesses and promote the just and efficient conduct of this litigation. All actions present common factual questions concerning the allegation that Google has monopolized or suppressed competition in online display advertising services in violation of federal antitrust law, whether that market is described singly as all display advertising services, as components of display advertising, or as some larger spectrum of digital advertising. Common factual issues in all actions include: (1) defining the relevant market for online display advertising services; (2) identifying the competitors in the market and their market shares; (3) the design and operation of Google's ad tech products and services, including alleged barriers to interoperability with competitors' products; (4) assessing the alleged anticompetitive effects of Google's conduct on market participants; and (5) Google's response to a competitive threat to its ad exchange known as "header bidding," which allegedly enabled publishers to use non-Google exchanges more effectively. Moreover, the 17 actions asserting Section 2 monopolization claims raise additional common factual questions, principally (1) Google's acquisitions of other digital ad tech companies, such as DoubleClick, and the competitive impacts of those acquisitions; and (2) alleged tying arrangements between, inter alia, Google's publisher ad server and Google's ad exchange. Sixteen of the 19 actions additionally assert that Google and alleged competitor Facebook entered into a secret agreement in 2018 to suppress the alleged "header bidding" threat to Google's market position.
In light of the concerns about the spread of COVID-19 virus (coronavirus), the Panel heard oral argument by videoconference at its hearing session of July 29, 2021. See Suppl. Notice of Hearing Session, MDL No. 3010 (J.P.M.L. July 12, 2021), ECF No. 103.
Header bidding allegedly involves code that publishers insert into the header section of their webpages that allows them to obtain bids from non-Google exchanges.
Centralization will promote the just and efficient conduct of the litigation by eliminating duplicative discovery and avoiding the risk of inconsistent rulings on pretrial matters, particularly on discovery disputes, Daubert issues, and dispositive motions. All actions, whether brought as putative class actions, individual actions, or governmental actions, will require common discovery from Google, which is the principal and common defendant, In addition, all cases will require discovery from Facebook because of the questions surrounding Facebook's status as a competitor; in at least 16 actions, discovery also will cover the Google-Facebook agreement. Third-party discovery will be significant, as the record indicates that there will be discovery concerning other alleged competitors, such as Amazon, as well as federal, state, and international investigations into Google's online display advertising practices. Few of the actions have commenced discovery, and those that have done so remain at a preliminary stage, making now an optimal time to structure the litigation to maximize efficiencies.
The actions allege that DoubleClick was the leading provider of publisher ad server tools in 2008, and that its acquisition by Google marked an important point in Google's dominance in online display advertising.
In opposition to centralization, plaintiffs primarily argue that (1) factual differences among the actions undermine Google's motion or at least warrant separate MDLs; (2) informal coordination and transfer under Section 1404 are practicable alternatives to centralization; and (3) exclusion of the State of Texas action is warranted because of the important governmental interests pursued by the states and the corresponding need for expeditious resolution of the action. After careful review of the record, we have determined that centralization of all 19 actions on the motion is warranted, despite the objections of the parties.
The assertion by plaintiffs in Cliffy Care and the 13 local newspaper actions that their antitrust claims concerning the Google-Facebook agreement are not appropriate for centralization is unpersuasive. The factual allegations concerning the agreement are substantially similar in all 16 actions that assert claims under Section 1 of the Sherman Act based on that agreement. Additionally, the absence of a Section 2 monopoly claim from some actions, like Cliffy Care, is not significant because the actions arise from a common factual core - Google's alleged suppression of competition in display advertising services. Indeed, the alleged "header bidding" threat to Google's ad exchange that the Google-Facebook agreement allegedly neutralized is at issue in all 19 actions. Lastly, the Cliffy Care plaintiffs' contention that relevant market issues are raised by the monopoly cases, but not by Cliffy Care, is an issue that remains to be determined in the litigation. Although plaintiffs assert that the agreement is a per se violation of Section 1 that requires no inquiry into the relevant market, defendants assert that recent Supreme Court precedent on Section 1 requires defining the relevant market to assess anticompetitive impacts, citing Ohio v. American Express, 138 S.Ct. 2274 (2018).
Moreover, the complaints allege that the agreement harms advertisers because Google and Facebook are the dominant players in display advertising, an allegation that on its own appears to raise market definition questions.
Separate MDLs for advertisers and publishers are not warranted. Many of the core factual issues in the publisher and advertiser actions overlap significantly, as discussed above - most importantly, defining the relevant market; assessing how Google's conduct affected the alleged components of the market; and how to calculate and apportion any damages between publishers and advertisers. Although advertiser and publisher actions also raise different issues - for example, Google's practices specific to ad tech tools for publishers versus tools for advertisers -"the transferee court may account, at his discretion, for any differences among the actions by using appropriate pretrial devices, such as separate tracks for discovery or motion practice." See In re Valsartan Prods. Liab. Litig., 433 F.Supp.3d 1349, 1352 (J.P.M.L. 2019).
Informal coordination is not a practicable and efficient alternative to centralization. The actions are pending in 16 districts, and involve seven distinct groups of plaintiffs' counsel. Third-party discovery will pose a further obstacle to informal coordination. Moreover, informal coordination appears inadequate to address the risk of inconsistent rulings in this factually and legally complex litigation. The possibility of transfer under Section 1404 also is not an efficient alternative given the number of involved districts. There are no pending Section 1404 motions, and none of the plaintiffs has indicated they would agree to transfer to a different district; rather, the record suggests that the vast majority will seek to stay in their chosen venues, hi the one action in which Google moved for Section 1404 transfer, the motion was denied. See State of Texas v. Google LLC, C. A. No. 20-0957, 2021 WL 2043184 (E.D. Tex. May 20, 2021). We pause to observe here that the factors governing centralization under Section 1407 differ from those governing transfer under Section 1404; thus, the order in State of Texas is not dispositive of our ruling on centralization. Moreover, the record before us is significantly different from the record in State of Texas - in particular, there are many more actions and districts at issue, as well as additional parties.
We further find that inclusion of the State of Texas action in this MDL is appropriate. The State Action raises the same factual questions as all the private actions concerning Google's alleged monopoly and the alleged Google-Facebook agreement to suppress competition. The states make no attempt to argue otherwise. Rather, they oppose transfer based largely on considerations relating to the relief requested, efficiency, and their sovereign status - primarily (1) they have decided not to pursue federal antitrust damages, as shown in their proposed second amended complaint, thus eliminating the overlapping federal damages issues raised by their action; (2) their action is more procedurally advanced, as it builds on an 18-month presuit state investigation that included voluminous discovery, whereas the other actions are in their infancy; (3) their sovereign enforcement action will not involve, and will be delayed by, the class certification proceedings in the class actions on the motion; and (4) as sovereigns, their choice of venue to vindicate the rights of their citizens and protect the public welfare should not be disturbed.
We are not persuaded that the asserted exclusion of federal antitrust damages from the State Action or the discovery posture disfavors transfer. Our decision to centralize is based on the common factual core shared by all actions, including the State Action - that Google has monopolized and suppressed competition in online display advertising services, including through the alleged 2018 agreement with Facebook. Eliminating one component of the request for damages does not alter that common factual core, which will require inquiry into complex fundamental issues such as defining the relevant market, identifying competitors, and assessing Google's alleged market power. Additionally, the State Action seeks relief that overlaps with the other actions in important ways even absent a request for damages under federal antitrust law -most importantly, declaratory and injunctive relief to enjoin the alleged unlawful practices by Google and Facebook, and structural relief that potentially could overlap with the structural relief sought by several publisher plaintiffs with respect to divestiture of one or more lines of Google's business.
The proposed second amended complaint in State of Texas makes the same factual allegations and asserts the same federal antitrust violations under the Sherman Act as the operative first amended complaint. Compare Proposed Second Am. Compl. at pages 10 to 119 with First Am. Compl. at pages 10 to 119.
The state plaintiffs still seek damages for violation of state antitrust and consumer protection laws.
The difference in the discovery posture of the State Action also does not weigh against transfer. The scheduling order in the action was entered only two months ago, and provides for a discovery period running through at least late 2022. That discovery, which is likely to be voluminous and complex, will overlap substantially with the discovery in the other actions on the motion, given that all actions raise the same core factual questions. Additionally, pretrial motions will address key issues affecting all actions like defining the relevant market - a matter that raises novel issues concerning the existence of a "two-sided" market, as well as differing proposed definitions by plaintiffs; thus, consistency in judicial rulings on these and other common issues will be important. Centralization will enhance the overall convenience of the parties, the witnesses, and the courts through coordinated proceedings on both discovery and motions in this MDL.
At oral argument, counsel for Google represented that no depositions have been noticed in the State Action.
See In re Watson Fentanyl Patch Prods. Liab. Litig., 883 F.Supp.2d 1350, 1351-52 (J.P.M.L. 2012) ("in deciding issues of transfer under Section 1407, we look to the overall convenience of the parties and witnesses, not just those of a single plaintiff or defendant in isolation").
We recognize the states' concerns regarding potential delay from centralization with putative class actions. But these are essentially case management concerns appropriate to raise with the transferee court for resolution. See, e.g., In re Ford Motor Co. DPS6 PowerShift Transmission Prods. Liab. Litig., 289 F.Supp.3d 1350, 1352 (J.P.M.L. 2018) (observing that concerns about "litigation delays" in the MDL were essentially case management concerns and "[i]t is incumbent upon the parties to bring their concerns to the attention of the transferee court and to propose ways to resolve them"), In the Panel briefing, Google has suggested that potential discovery delays from class certification proceedings could be managed by first scheduling a common discovery period, and then having the class claims proceed to class certification while the non-class claims proceed to dispositive motions. We encourage the parties to collaborate on these and other proposals to streamline pretrial proceedings and to bring their proposals to the attention of the transferee court. As with any MDL, the degree of consolidation or coordination is a matter soundly dedicated to the discretion of the transferee judge. See In re Hyundai and Kia Fuel Economy Litig., 923 F.Supp.2d 1364, 1365 (J.P.M.L. 2013).
See Google Reply Br., Doc. No. 89, at 3 (J.P.M.L. June 2, 2021).
The state plaintiffs' status as sovereigns does not weigh against inclusion of their action. The Panel regularly has transferred state enforcement actions to antitrust MDLs (and to other types of MDLs), based on the traditional Section 1407(a) criteria, rejecting the same kinds of arguments the state plaintiffs are making in this docket. See, e.g., In re Generic Pharmaceuticals Pricing Antitrust Litig., MDL No. 2724, 2017 WL 4582710 (J.P.M.L. Aug. 3, 2017) ("the Panel has transferred state enforcement actions to MDLs involving cases brought by private litigants with some regularity"); In re Auto Body Shop Antitrust Litig., MDL No. 2557, Doc. No. 306, at (J.P.M.L. Dec. 12, 2014), ECF No. 306, at 1-2 ("[The State of Louisiana] argues that its enforcement action should proceed separately from the private actions, This argument is unconvincing, as we often have transferred state enforcement actions to MDLs that involved cases brought by private litigants."). Putting aside our past decisions centralizing state actions with private actions, inclusion of the State of Texas action is especially compelling here in light of the substantial overlap in the alleged Google conduct, plaintiffs' overlapping but different proposed relevant markets, and the involvement of third-party discovery. Additionally, the states' arguments to exclude their action rely in large part on proposed legislation that, if enacted, would effectuate that desired outcome. However, we must apply the law currently in effect, without speculating about what future legislation might be passed.
See State Antitrust Enforcement Venue Act of 2021, H.R. 3460 and S, 1787, 117th Cong. (2021).
II.
As to State of Texas, Google also requests that we order transfer of the action for trial under Section 1407(h), which authorizes the Panel to transfer "for both pretrial purposes and for trial, any action brought under section 4C of the Clayton Act" - i.e., State parens patriae actions seeking monetary relief for federal antitrust violations. Google's request remains live, notwithstanding the states' proposed second amended complaint dropping their request for damages under Section 4C, because the operative complaint currently in effect asserts Section 4C as a basis for jurisdiction and relief. However, we have determined that the record does not contain adequate information to enable us to evaluate whether transfer for trial is warranted. In this litigation, we believe that the determination of transfer for trial is better made after the issues in the actions are more fully developed, presumably after threshold motions have been resolved and core common discovery completed. Furthermore, we anticipate that the transferee court, with the benefit of those proceedings and briefing by the parties, will make a suggestion on whether conducting the trial of the State Action in the transferee district would serve the convenience of the parties and witnesses and promote the just and efficient conduct of the litigation. The states' recent filing of a motion for leave to file a second amended complaint that, if granted, would delete reliance on Section 4C, only underscores the need for further development of the issues in these actions before a decision on transfer for trial can be made. Accordingly, we deny without prejudice Google's request to transfer the State Action for trial under Section 1407(h).
Google recently advised the Panel that it has reserved its right to oppose the amendment and continues to seek transfer of the State of Texas action for trial under Section 1407(h). See Google's Response to Supplemental Information, Doc. No. 117 (J.P.M.L. Aug. 6, 2021).
We anticipate that transfer of State of Texas for trial, if appropriate, would take place through remand to its originating court under Section 1407(a) and simultaneous transfer back to the transferee court under Section 1407(h) for trial. See In re Disposable Contact Lens Antitrust Litig., MDL No. 1030, Doc. No. 81, Order Remanding Actions under 28 U.S.C. 1407(a) and Transferring Actions under 28 U.S.C. 1407(h) (J.P.M.L. Dec. 1, 1998). Thus, the party seeking transfer of the action for trial should first file a motion with the transferee court seeking a suggestion of remand and simultaneous transfer for trial under Section 1407(h), consistent with our rules. See Panel Rules 10.2 and 10.3. In the event that that transferee judge declines such a request, any party to the State of Texas action may file directly with the Panel a motion seeking remand and simultaneous transfer for trial under Section 1407(h).
III.
We conclude that the Southern District of New York is an appropriate transferee district for this litigation. The record indicates that the advertising and publishing industry around which these actions revolve have a strong presence in New York, where the Associated Newspapers action is pending. Two organizations representing news entities and other online content providers filed amicus briefs supporting centralization of the private actions in the Southern District of New York, and plaintiffs in 14 actions request this district in the event that centralization is granted over their objection. Facebook indicated at oral argument that the Southern District of New York is an appropriate alternative to its preferred California forum. Moreover, the record indicates that significant Google operations concerning the issues in this litigation are located in New York and that much of the common evidence is there as well.
We believe that assignment of this MDL to an experienced transferee judge is needed, considering the complexity of the factual and legal issues, the anticipated breadth of third-party discovery, and the parties' concerns over case management efficiencies in litigation of this scope. Judge P. Kevin Castel, to whom we assign this litigation, has presided over three MDLs, and has the willingness and ability to manage this litigation efficiently. We are confident he will steer this litigation on a prudent course.
IT IS THEREFORE ORDERED that the actions listed on Schedule A and pending outside the Southern District of New York are transferred to the Southern District of New York and, with the consent of that court, assigned to the Honorable P, Kevin Castel for coordinated or consolidated pretrial proceedings.
IT IS FURTHER ORDERED that MDL No. 3010 is renamed In re: Google Digital Advertising Antitrust Litigation.
SCHEDULE A
(Image Omitted)