Last Friday, October 22, a coalition of US States led by Texas released an unredacted version of the amended complaint filed against Google over its ad tech business, after a federal judge in New York held that much of the antitrust suit could be unsealed.
The original complaint filed in December 2020, echoes the findings of numerous competition authorities, including the UK Competition and Markets Authority, the French Autorité de la concurrence, and the Australian Competition and Consumer Commission (more on this later), all of which have found that Google has used its market power across the ad tech supply chain to engage in a variety of leveraging tactics that have distorted competition. In the meantime, the US Department of Justice is reportedly preparing to file its own lawsuit against Google, while the European Commission is also probing the latter over its ad tech practices.
By way of reminder, ever since its acquisition of DoubleClick, Google has become the largest ad tech vendor across each step of the value chain, with market shares as high as 90-100%, while also a major publisher itself (selling inventory on its owned and operated properties like YouTube). While most of this was more or less known, the Texas complaint made headlines for claiming that when faced with the prospect of Facebook supporting a disruptive technology known as Header Bidding, which Google viewed as an “existential threat,” Google struck a deal with Facebook. According to the so-called Jedi Blue agreement, Facebook would curtail its Header Bidding initiatives in return for special privileges when bidding in Google’s auctions.
While originally a significant portion in the complaint was redacted, we now have access to the full lawsuit thanks to the order of the US District Court. Most of the previously redacted passages are (rather juicy) quotes from internal Google and Facebook documents, as well as information on Google’s fees. As such, the unredacted complaint does not include anything new in terms of the claims made against Google, or the theories of harm advanced by the US States. Even so, the unredacted lawsuit shows that these theories of harm are supported by a fairly wide range of damning internal documents. Internal evidence is important, as it makes it much harder for the target of the complaint to raise arguments that are directly contradicted by its own views as expressed internally in tempore non suspecto.
Indeed, if there is a common theme running across the various internal documents cited in the unredacted lawsuit, this is the inconsistency, or even stark contradiction, between Google’s public position on a variety of issues and the views of its own employees and executives. With that in mind, I will now go through what I consider the “hottest” internal evidence cited in the unredacted complaint, grouped together according to the subject to which it pertains.
On Header Bidding and Jedi Blue
While publicly stating that it did not see Header Bidding as a “threat” to its business, internal documents show that Google viewed Header Bidding as nothing short of an “existential threat.” By way of background, Header Bidding was invented by publishers and rival ad tech vendors in reaction to Google’s favoritism towards its own ad exchange, which costed publishers real money, as internally admitted by Google employees. As Header Bidding rose to prominence, Google employees discussed in October 2016 “options for mitigating growth of header bidding infrastructure.” When one Google employee proposed the “nuclear option” of reducing Google exchange fees down to zero, another responded that the problem with competing on price is that it simply “doesn’t kill HB [header bidding].” In response, Google developed its own Header Bidding-like solution now called Open Bidding (code-named “Jedi”), whose success it measured not by financial targets or output increases, but by how much it stopped publishers from using header bidding. However, one senior Google employee commented that Jedi “generates suboptimal yields for publishers and serious risks of negative media coverage if exposed externally.”
Meanwhile, Google was alert to the risk of large entrants supporting Header Bidding. In an internal 2016 presentation to senior Google executives, a Google employee expressed concern about Amazon, Criteo, and Facebook enabling the growth of Header Bidding, stating “to stop these guys from doing HB we probably need to consider something more aggressive.” The presentation plainly asserted that Google’s “goal/mandate” was to “[f]orestall major industry investment in HB & HB wrapper infrastructure.” In response to Facebook signalling its support for Header Bidding, a Google executive impressed in a company deck the “Need to fight off the existential threat posed by Header Bidding and FAN. This is my personal #1 priority. If we do nothing else, this need[s] to [be] an all hand[s] on deck approach.”
Google eventually invited Facebook to the negotiating table. According to the lawsuit Google promised Facebook a series of advantages when bidding on Open Bidding (where Google would still retain control and charge rivals a fee), in return for Facebook curtailing its Header Bidding initiatives. These special terms included speed advantages and assistance in identifying users (which is crucial for online ad auctions). In an internal document Google memorialized that “FAN requires special deal terms, but it is worth it to cement our value.” (Meanwhile, in its web support manager Google states that all auction participants compete “equally”). Any doubt as to the understanding of Google and Facebook when entering into the Jedi Blue agreement (signed by top executives Philipp Schindler of Google and Sheryl Sandberg of Facebook), is dispelled by the internal evidence cited in the complaint. In internal communications, Facebook executives noted that “They [Google] want this deal to kill header bidding.” According to documents cited in the complaint, Facebook “believed strongly” that partnering with Google was “relatively cheap compared to build/buy and compete in zero-sum ad tech game.” When considering whether to sign the deal, the team outlined that Facebook had four options: to “invest hundreds more engineers” and spend billions of dollars to lock up inventory to compete, exit the business, or do the deal with Google. Facebook opted for the latter, agreeing to minimum quotas in terms of bidding and winning in Open Bidding auctions.
Now, while we still do not know all the facts, Google’s deal with Facebook could well be considered anticompetitive by object under EU law, as its (apparent) object was to substitute the risk of competition for practical cooperation. According to the lawsuit, Google essentially bought off Facebook – in this respect the Jedi Blue deal reminds me of pay-for-delay agreements in the pharmaceutical industry.
On Google’s ad exchange and its fees
In response to the ACCC’s interim report on ad tech, Google filed a supplementary submission arguing that online ad exchanges (which are unregulated) cannot be compared to stock exchanges (which are heavily regulated to prevent e.g., conflicts of interests from arising). However, the unredacted lawsuit shows how Google’s own employees frequently compare AdX with a stock exchange, making remarks rather unfavourable to Google. According to an internal document, “[h]undreds of thousands of publishers and advertisers use [Google’s] AdX [exchange] to transact inventory, and more daily transactions are made on AdX than on the NYSE and NASDAQ combined.” At the same time, Google owns the largest buyers and sellers transacting on its exchange, giving rise to clear conflicts of interests, as found by the ACCC. According to the lawsuit, one senior Google employee admitted that “[t]he analogy would be if Goldman or Citibank owned the NYSE.” Needless to say, this would be unthinkable in the financial world.
Besides highlighting the clear conflicts of interest Google faces, the comparison with stock exchanges is useful in illustrating the very high fees Google charges. As admitted by one Google employee, “an exchange shouldn’t be an immensely profitable business” like Google’s AdX, but should instead be like “like a public good used to facilitate buyers and sellers.” According to the lawsuit, AdX charges a fee between 19-22%, which is two to four times higher than the fees of competitors. Google recognized internally that “20% for just sell-side platform/exchange isn’t likely justified by value.” In an internal email from 2017 – when Header Bidding had spread like wildfire across publishers and threatened to disintermediate Google – a Google employee expressed the view that exchange “margins will stabilize at around 5 percent. Maybe it will happen by this time next year or in early 2019. This creates an obvious dilemma for us. AdX is the lifeblood of our programmatic business. … What do we do?” Yet Google has been able to maintain the same high revenue share – and according to the decision of the Autorité de la concurrence (which Google agreed not to contest in court) this is the result of anticompetitive conduct (Google leveraging its publisher ad server to advantage AdX).
As for Google’s overall fees (across the value chain), these range between 30 to 42% according to the lawsuit, with the buy-side fees of Google Ads not transparent to advertisers. In a discussion between Google employees about Google Ads’ fees, one employee asked: “Buyers don’t know that [we] take a 15 percent fee? I didn’t realize that.” Another clarified that the fee “is not transparent.” Overall, these high fees are likely to reflect Google’s market power and the absence of competition. As both the CMA and the ACCC found, ad tech fees would likely be lower in a competitive market.
Google states in its communications to the public and regulators that it takes our privacy very seriously. At its I/O 2021, Google spent part of the keynote talking about privacy and security. As 9To5Google reported, Google followed up with a “We protect your privacy” message in Gmail and Google Photos. As of lately, Google cites privacy as the all-purpose justification for various problematic practices, including cutting third-party DSPs’ access to YouTube inventory. A cynic, of course, would take these statements with a grain of salt, considering that the concept of privacy is fundamentally at odds with the very business model of Google. Still, I felt rather uncomfortable reading in the unredacted complaint how Google apparently took credit for slowing down and delaying the EU’s ePrivacy Regulation in collaboration with other tech companies. The quote appears in a July 2019 internal memo prepared in advance of a meeting with other major big tech companies.
Conclusion – the ACCC report
Judging by the internal evidence in the unredacted complaint, it comes as little surprise that Google did not wish for these quotes to be released in the public domain. Against the wealth of evidence cited in the complaint, defending the lawsuit in court will be no easy feat for Google. The case is expected to go on trial in 2023.
Meanwhile, regulators across the world continue to make damning findings with respect to Google’s ad tech business. The latest news comes from Australia, where last month the ACCC released its final report on its ad tech inquiry. Confirming the earlier findings of its interim report (see summary here), the ACCC concluded that competition for ad tech services in Australia is ineffective, with Google dominating the supply of ad tech services. In line with overseas regulators, the ACCC held that Google’s vertical integration gives rise to clear conflicts of interest and has allowed Google to engage in various leveraging practices that have restricted competition. Interestingly, the ACCC has recommended that new ex ante rules be introduced with a view to promoting competition and transparency in the provision of ad tech services. These rules could form part of a wider ex ante regime for digital platforms (which seems to be to some degree inspired by the Strategic Market Status regime in the UK). The ACCC’s key recommendation is that it be given the power to introduce ex ante rules to address Google’s conflicts of interest and competition issues in the ad tech supply chain. Another noteworthy recommendation is that the ad tech industry should develop standards that require all ad tech providers (not only Google) to be transparent on their fees and take rates, failing which the ACCC would have the power to step in and develop specific rules to improve transparency. Now, let’s see how the government reacts to the ACCC’s recommendations, and whether Australia will follow the UK by setting an ex ante regime tailored to digital platforms. But, if anything seems certain, this is that Google’s ad tech business will still be in the spotlight for years to come.
[Disclosure: Damien and I acted for the lead complainant in the French case against Google, and we have advised various parties over issues related to online advertising and Google’s ad tech practices. All views in this post are mine and should not be attributed to any client]