The Anticompetitive Effects Arising From Google’s Conduct

VIII. ANTICOMPETITIVE EFFECTS ARISING FROM GOOGLE’S CONDUCT

971. “[C]ompetition matters because it determines the outcomes in markets,” i.e., the quality and price of the products and services available to (1) consumers, in the case of general search services, and (2) advertisers, in the case of the ads markets. Tr. 5774:7–13 (Whinston (Pls. Expert)).

972. Unlike foreclosure, competitive effects are ideally estimated relative to a but-for world. Tr. 5774:14–5774:25 (Whinston (Pls. Expert)). But, here, it is impossible to know what a but-for world would look like. Id. 5775:1–5776:21 (A but-for world is impossible to determine; there are too many unknowns about what would have happened but for Google’s behavior.). There are many uncertainties about what market participants, including Google, and potential market participants would have done but for Google’s exclusionary contracts. Id.

For example, Google might have entered into unconditional revenue sharing agreements, Google might have entered most-favored supplier contracts, or Google may not have entered into any search distribution contracts. Id. 5775:1–5776:5. It cannot be known how, in response to changes in Google’s conduct, rivals might have changed their search investments and what impact those investments would have had on search quality. Id. 5776:6–21. Rivals may have expanded, and new rivals might have entered. Id.

973. Determining what the but-for world would have looked like in this case would require a “time machine into a world that doesn’t exist.” Tr. 5775:1–5776:5 (Whinston (Pls. Expert)). Because of these uncertainties, a competitive effects analysis cannot yield a precise quantitative answer. Id. 5776:6–21 (“[I]t’s just an impossibility.”).

Instead, a Section 2 competitive effects analysis will necessarily be a qualitative exercise examining how competition may have changed if Google’s exclusionary contracts were not in place for a decade. Id. 5776:22–5778:3 (It is possible to consider less restrictive alternatives and how those less restrictive alternatives change “the fundamental forces in the market,” which in turn “would change market outcomes.”).

974. There are many possible alternatives that would have been less restrictive than Google’s contracts, two of which are: (1) unconditional revenue share payments and (2) most-favored supplier agreements. Tr. 5776:22–5777:14 (Whinston (Pls. Expert)). Both of these alternatives would have created a “more equal playing field in terms of [general search] distribution.” Id. 5777:15–5778:3.

975. In his assessment of competitive effects, Prof. Murphy disregarded a wide range of outcomes that could have arisen but for Google’s contracts. For example, he did not analyze the impact of a rival gaining a single exclusive default on Android or Apple devices because he assumed this would not happen, even in the absence of Google’s contracts. Tr. 10013:14– 10014:3 (Murphy (Def. Expert)) (On Android, “I don’t believe they would.”); id. 10027:8–13 (On iOS, he had no opinion on what share rivals would gain if they had exclusivity because he “[did not] believe that would happen in the but-for world.”).

976. Prof. Murphy also disregarded the impact of any potential new entry, including by Apple or Branch, because he assumed that no additional entry would have occurred absent Google’s contracts. Tr. 10047:22–10048:10 (Murphy (Def. Expert)) (“I don’t think there would have been additional entry, no.”); id. 10016:3–16, 10022:9–15 (did not consider the impact of Apple entering search because he “didn’t see why that would be a relevant question here” and it “wasn’t important for what [he] was doing”); id. 10040:4–24 (did not consider the impact of Branch entry because “[i]t came up late in the case” and Branch “does something different”); id. 10040:25–10042:20 (no opinion about the likelihood of Branch entering general search).

977. Google’s search distribution contracts harm competition in the general search services, Search Ads, and Text Ads markets in the United States in at least two ways: (1) the contracts prevent rival GSEs from accessing scale, weakening them as competitors; and (2) the contracts reduce the incentives to compete on quality and price for Google, current rivals, potential entrants, and distributors. Tr. 5781:8–5782:8 (Whinston (Pls. Expert)). This reduced competition reduces the quality and options available to consumers in all three markets. Id. 5854:11–20).