Testimony of Brad Smith, Senior Vice President and General Counsel of Microsoft before the United States Senate Judiciary Committee Subcommittee on Antitrust, Competition Policy and Consumer Rights
Hearing on The Google/Yahoo! Agreement and the Future of Internet Advertising
July 15, 2008 – Chairman Kohl, Ranking Member Hatch, and honorable members of this Subcommittee, my name is Brad Smith, and I am Senior Vice President and General Counsel of Microsoft Corporation. Thank you for the opportunity to provide Microsoft’s perspective on these important issues. Before going into detail on the topics of search advertising and the agreement between Google and Yahoo!, which together control approximately 90 percent of search advertising, I thought I would begin by providing an overview of Microsoft’s perspective on the deal.
I. Summary of Why the Google-Yahoo! Deal Would Reduce Choice and Innovation and Lead to Higher Pricing
Less than a year ago, I appeared before the Congress to discuss the serious challenges facing the state of competition in the critical space of online advertising. In the intervening months, I regret to report that the state of competition clearly has not improved.
Although the online world is still in its relative infancy, we find ourselves at a crossroads. Competition in search and search advertising is critical to our society and economy. The issues before this Subcommittee today could very well shape the extent to which the Internet continues to develop into a thriving marketplace of commerce and ideas.
Specifically, we are here today because Google and one of its chief rivals, Yahoo!, have teamed up in a deal that affects approximately 90 percent of all search advertisements sold in this country. If permitted to proceed, we believe the Google/Yahoo! agreement would effectively create a monopoly in search advertising — to the extent one does not already exist — and further reduce competition.
To be clear, obtaining a high market share through ingenuity and hard work is how our system rewards those businesses that succeed in a competitive marketplace. Indeed, Microsoft is quick to recognize that a company like Google has worked hard to achieve many of its great accomplishments. However, compared to earned success, achieving or entrenching dominance through collusive agreements or other artificial means undermines competition.
Microsoft believes the Google/Yahoo! deal harms competition in several critical ways. Advertisers and online content providers would be harmed through price coordination that will establish higher prices and limit choice. Consumers would be put at risk as Google expands its ability to collect the personal information of users passing through its search gateway. On an even more fundamental level, Google’s monopoly power would increase its ability to shape what people get to see and experience online.
Ultimately, the long-term ramifications of this deal would undermine the very diffusion of power the Internet is supposed to promote. There is the very real prospect that we would regress to the days when information and communications were in the hands of only a few national, broadcast television companies — a state of affairs antithetical to the Internet’s core purposes, principles, and promise.
Before addressing Microsoft’s specific concerns with the deal, it is also worth pausing to recognize why we are here today. The central question faced by this Subcommittee and regulators is not whether this deal is good or bad for Yahoo!’s search advertising business. The antitrust laws exist to protect competition, not individual competitors. Thus, the central question is whether the deal between Google, which accounts for about 70 percent of the search advertising market, and its main search rival Yahoo!, with around 20 percent market share, would harm competition.
Moreover, although we would welcome a discussion about the pro-competitive benefits of a Microsoft/Yahoo! transaction at the appropriate time, it is little more than an academic question at this stage. The Google/Yahoo! agreement is the one that Yahoo! has chosen to pursue. It is the Google/Yahoo! deal that is under investigation by the United States Department of Justice and numerous state attorneys general, and it is the Google/Yahoo! deal that, we believe, poses an immediate and substantial threat to competition.
II. Competition in Search and Advertising is Critical to our Society and Economy.
A. Online Search Advertising Is Critically Important to the Internet and the Businesses that Use It to Reach Consumers.
Search engines are a critical gateway to the Internet and its almost limitless content and services. An estimated 65 percent of online shoppers conduct product research using search engines. Search engines enable us to find the content we need and to harness the full power of the web. They are, therefore, critical to the growth and health of the Internet and to our broader economy, particularly as the Internet continues to grow as a destination for information, communication, and commerce. In May 2008, for example, there were over 7.8 billion search queries conducted in the United States.(2)
Online advertising is the “fuel” that drives all the Internet has to offer. Advertising, for example, is what enables search engines to provide their important services for free. It also enables content providers on the web, such as MSNBC.com, ESPN.com, and FoxNews.com, to offer free content and services. Online advertising expenditures likely will exceed $27 billion in the United States in 2008 and are expected to grow to about $42 billion by 2011.(3)
Search advertising, in particular, offers a unique value proposition to advertisers because it allows them to target ads to the real-time desires of a user. And the advertiser typically only pays for the ad if the user actually clicks on it. This one-two punch provides compelling value that, at least today, allows small, medium, and large businesses alike to succeed using the power of search ads. In 2007, approximately $8.6 billion was spent on search advertising in the United States.(4)
B. Increasingly, One Company Controls the Search “Gateway”: Google.
Google operates the dominant search engine and is the dominant provider of search advertising in the United States. It accounts for approximately 70 to 75 percent of search advertising revenue (and roughly the same number of search queries).(5) Yahoo! is Google’s next closest search advertising competitor with around 20 percent market share. Microsoft is a distant third with less than 10 percent. In just 10 years or so, Google has amassed a market capitalization of around $166 billion, nearly all attributable to its search advertising business. To put this in context, Google’s market capitalization is as big as Coca-Cola and Boeing combined.(6)
Last year, the FTC put to rest any doubt as to whether Google controls the search gateway in its decision on the Google/DoubleClick transaction. Following its lengthy and thorough investigation, which involved a review of Google’s internal documents (likely at least a million pages of Google’s own documents) and discussions with scores of advertisers and industry participants, the FTC reached an important conclusion: Google is “the dominant provider” in the search advertising market.(7) As a fundamental matter, the FTC concluded that other forms of advertising (online or offline) should not serve as a basis to eliminate concerns about Google’s dominance in search.
C. Overview of Search Advertising.
1. The Mechanics of Search Advertising.
Search advertising refers to the text links that typically appear at the top and right side of search engine results pages. Search engines interact with a search advertising platform. Both technologies work together to provide a seamless experience to the user, but work differently behind the scenes and play very different roles in this important ecosystem.
After we enter a query into a search engine, we actually see two different types of results on the same page. The first are the search query results themselves, also called “algorithmic” search results. These unpaid results are generated by the search engine’s index of online content and attempt to list the websites the engine decides are most relevant to the user’s search query.
The second type of search results are links to paid advertising that typically appear just above or to the right of algorithmic search results and are delivered by the search advertising platform. Advertisers typically pay only when users click on these textual ads.
Although search ads appear to be quite simple, offering them is a huge and complex business. Operating a search engine and a search advertising platform can require billions of dollars to maintain the necessary infrastructure. That is why there are just three companies participating in this market today — Google, Yahoo!, and Microsoft. When it comes to search advertising, alternatives are not just one click away. This is a tough business in which to succeed. It is a business where size and “scale” make all the difference. Thus, the bigger the leader gets, the harder it is for everyone else to compete.
2. Google has the ability to influence and manipulate pricing of search advertising.
Google, Yahoo!, and Microsoft use a process to determine which search ads to place and in which order. This process involves an auction among advertisers. Google has suggested that it does not influence pricing of search advertising because it determines the prices through this auction.(8) Contrary to Google’s claim, however, the advertiser auction on its system is no ordinary auction where buyers and sellers set price without active intervention of the auctioneer. Rather, Google has the ability to influence and determine prices through various means, such as minimum prices and secretive “quality scores.” In fact, in the Google auction, the highest bidder in the keyword auction may not win, and what an advertiser “bids” is not necessarily what an advertiser actually “pays.” Indeed, many advertisers have complained about the lack of transparency in Google’s auction process.(9)
III. The Google/Yahoo! Agreement.
A. Overview of the Agreement.
The events leading up to this hearing are well known. Namely, there exist serious and important questions regarding the effect of the June 12, 2008 advertising agreement between Google and Yahoo!. Under the agreement, Yahoo! will be able to place search and contextual advertising from Google on Yahoo!’s properties in the United States and Canada. In exchange, Google has agreed to give Yahoo! an undisclosed portion of the revenue from Google’s ads shown on Yahoo!’s search pages. The deal has an initial four-year term and can last as long as ten years.(10)
Google and Yahoo! have not disclosed their agreement to the public.(11) Instead, they have provided selected details in public statements and mandatory filings with the Securities and Exchange Commission (“SEC”).(12) While Google and Yahoo! have suggested that Yahoo! is not obligated to place any Google ads next to Yahoo! search results, they are silent on whether there are limits to the amount of advertising from Google that Yahoo! could substitute for its own. In fact, there appear to be several incentives for Yahoo! to hand over more, not less, of its advertising sales to Google.(13)
Yahoo!’s President Sue Decker suggested during the company’s June 12 investor call that Yahoo! will rely on whichever search ad(s) will fetch a higher price (i.e., “better monetize”). In essence, therefore, the deal effectively creates a “price floor” (i.e., Google’s price) because there is no incentive for Yahoo! to sell its ad space inventory for less than Google’s platform can realize.(14)
Yahoo! projects that the deal will generate additional operating cash flow of as much as $800 million per year from Google. This additional money presumably will flow from the pockets of advertisers — in many cases in the form of higher prices for the same ads they purchase directly from Yahoo! today.
Google and Yahoo! have not touted any benefits from operational integration between the companies.(15) This suggests that the Google/Yahoo! agreement is a bare marketing and pricing agreement between head-to-head competitors.(16)
Both Google and Yahoo! have presented the deal as “open,” although it hardly seems to be so in any practical way. In the software industry, the word “open” is often synonymous with “interoperability,” which is not going to occur as a result of this agreement. Notably, it is a “one-way” agreement — Google can sell search ads on Yahoo!, but Yahoo! cannot sell search ads on Google. Moreover, there is no reason to believe the same arrangement will work in practice for any non-Google competitor, as Google’s typically higher auction prices invariably will trump the lower prices of Google’s competitors.
Finally, Google has represented that this kind of arrangement is commonplace in many industries and has compared the deal to Toyota selling its hybrid engine technology to rival General Motors. This analogy simply does not hold water. Google is not selling Yahoo! a part (analogous to a hybrid engine) that Yahoo! needs to build a product (a completed car). Google is replacing Yahoo! as the seller of the product itself — search advertising. If anything, it is more like Toyota selling GM the whole car and the two companies agreeing to sell Toyotas in GM showrooms in instances where Toyotas can fetch a higher price! The reality is that GM and Toyota could no more enter into such an agreement than Google and Yahoo!. This is anything but a standard supply agreement.
IV. The Google-Yahoo! Deal Would Reduce Choice and Innovation for Advertisers, Content Creators, and Consumers and Lead to Higher Pricing.
The Agreement Entangles the Dominant Provider of Search Advertising with Its Closest Competitor.
The Google/Yahoo! agreement contemplates significant, ongoing coordination between the dominant provider of search advertising and its chief rival. Together, Google and Yahoo! control an estimated 90 percent of search advertising, with Google alone accounting for over 70 percent. With search advertising fueling the anticipated doubling of online advertising over the next three years, the deal could make Google bigger than any newspaper chain or any television network and provide Google the largest concentration of advertising control in history.(17) The growing importance of the Internet, and media in general, makes this concentration of power all the more concerning.
B. The Agreement Harms Competition in Several Ways, All of Which Would Further Entrench Google’s Dominance in Search and Reduce Choice.
Microsoft believes that the Google/Yahoo! agreement likely violates the antitrust laws in several fundamental ways. Some of the anticompetitive aspects of the agreement are discussed below.
First, the agreement reduces Yahoo!’s incentive to compete with Google. Because Yahoo! will share in Google’s revenue, it will make more money when Google makes more money. Similarly, Yahoo! will make less money when Google makes less money. Thus, Yahoo!’s incentive to compete against Google following the deal will be reduced. In fact, any competitive decision that Yahoo! makes regarding pricing, innovation, or capital expenditures will have to be weighed against the revenue that Yahoo! would receive from Google’s higher profits. The antitrust laws have long disfavored agreements that result in incentives for competitors to “pull their punches” with one another. The concerns are particularly acute in instances where the two companies involved in the coordination account for such an enormous percentage of the market.
Second, this agreement increases the probability that Yahoo!’s search advertising platform will enter a downward spiral that will further weaken its competitiveness, particularly as Yahoo! funnels more search share to Google and as advertisers realize they can no longer get “discounts” on Yahoo!’s platform. In turn, this would reduce even further Yahoo!’s incentive to invest and compete in search advertising and would result in an even greater dependence on Google. Interestingly, an internal Yahoo! document from this past January contemplated some of these same effects when Yahoo! suggested that the deal before this Subcommittee today would be harmful to Yahoo! and result in Google having an “effective monopoly.”(18)
Third, the agreement effectively establishes a per se illegal price floor. In explaining the deal, Yahoo!’s President stated that Yahoo! will rely on Google’s ads when those ads command a higher price (i.e., “better monetize”). Quite simply, Yahoo! and Google are setting a “price floor” (i.e., Google’s price) for ads on Yahoo!’s properties. Yahoo! will never have an incentive to sell an advertisement for less than Google is offering.
The legal precedent in this area is straightforward. In United States v. Socony-Vacuum Oil Co., a case taught in law schools across the country as “black-letter law,” the Supreme Court established that agreements between competitors that stabilize price or establish a “price floor” constitute per se illegal price-fixing under the antitrust laws — regardless of justifications, intent, or the mechanism used to effectuate the fix.(19) In that case, large gasoline refiners effectively created a price floor for gasoline by agreeing to buy from smaller refiners cheaper “distressed” gasoline that otherwise would have reduced prices.
Similar to the situation in Socony-Vacuum Oil, Google is effectively agreeing to buy up Yahoo!’s ad space wherever that ad space would otherwise be available at lower prices. As a result, keywords that cost advertisers less on Yahoo!’s platform before the deal would no longer be available, because those advertisers would have to pay the higher price at which those keywords are sold by Google in order to advertise on Yahoo!. Put another way, advertisers would no longer get the “bargains” they have enjoyed as a result of competition between Yahoo! and Google.
C. The Loss of Competition From the Agreement Would Harm Consumers, Advertisers, and Content Providers in Several Ways.
The reduction in competition resulting from the agreement would have important and longstanding effects on the search marketplace, those that rely on it, and the Internet as a whole. For example:
The agreement will reduce choice and place even more power and control over a major industry and online content in Google’s hands. Google will control even more of consumers’ personal information. A Google-Yahoo! partnership not only gives Google the ability to place advertisements next to nearly 90 percent of online search results — it also gives Google access to up to 90 percent of all search queries and online behaviors, including the sites consumers click on and, in some cases, what they do on those sites. Similarly, with weaker competition, one has to wonder what would constrain Google’s already questionable privacy practices relating to the reservoir of personal data it holds. Even today, Google has shown a willingness to push the line when it comes to privacy. Indeed, only with substantial prodding did Google finally add a privacy link to its home page.
Content providers, which rely on neutral and competitive search in order to reach users through advertising and unbiased search results, will become more beholden to Google’s power. Advertisers also will find themselves left with no choice but to place ads with Google, despite its non-transparent policies.
The agreement will result in higher prices for advertisers and content creators. As already discussed, the agreement effectively establishes a minimum price floor and otherwise reduces the incentives of Google’s closest competitor to compete. The reduction in competition between Google and Yahoo! may also mean that web content creators would receive less money from Google or Yahoo! for placing toolbars and ads on their sites, which is an important source of revenue for many online businesses.
D.The “Non-Exclusive” Label Does Not Immunize the Agreement from the Antirust Laws.
Finally, both Google and Yahoo! have presented the deal as “open” and “non-exclusive” in the sense that it does not prohibit Yahoo! from running ads served by its own search advertising platform or third parties. Even putting aside the significant incentives built into the agreement to funnel ads to Google, such labels carry little weight. Antitrust courts and regulators focus on the substance and ultimate effect of agreements, rather than their form or labels. In their own guidelines (and enforcement history), regulators recognize that non-exclusive arrangements can violate the antitrust laws if they result in one firm maintaining or obtaining monopoly power, or, more generally, if the arrangement results in conditions that are harmful to competition.(20)
I am fully aware that the presence of Microsoft at this hearing must strike some as ironic, given our own antitrust history. There also is no doubt that we have an interest here as a competitor to Google and Yahoo!. Given the concerns that have been expressed by others, the Department of Justice will hear from several advertisers, agencies who place advertising on behalf of businesses, consumers, and, of course, the people who create content and services that are freely available on the Internet. Already we have heard from advertisers — many of them Google’s own customers — about the negative impacts of the company’s deal with Yahoo!.(21)
Ultimately, it is well recognized that Google could not acquire advertising space from Yahoo! as part of an acquisition. When it comes to the issues before this Subcommittee, Google should not be allowed to achieve an outcome through an agreement that it would not be permitted to achieve otherwise. The effect of this agreement would be to further entrench the control of the dominant supplier of search advertising and, in the process, reduce choice and innovation and increase prices. As a result, we believe the Google/Yahoo! agreement raises some very serious concerns and may very well be illegal under the antitrust laws.
Mr. Chairman and members of the Subcommittee, thank you for the opportunity to appear before you today.
iCrossing, How America Searches: Online Retail at 2 (Sept. 24, 2007).
Nielson Online, May U.S. Search Share Rankings (June 19, 2008), available at
EMarketer, Search Marking, the Behemoth Online Advertising Format (Feb. 2008), available at
Id. (estimating that Google raked in 75 percent of U.S. paid search advertising in 2007); comScore qSearch, All qSearch Properties media category (May 2008) (with Microsoft analysis) (listing Google’s U.S. query share, along with its partners, at 75 percent in May 2008; noting that Yahoo!’s query share in May 2008 was 17 percent and Microsoft’s was 7 percent).
Drake Bennett, Stopping Google, BOSTON GLOBE (June 22, 2008), available at
Statement of the FTC Concerning Google/DoubleClick at 3, FTC File No. 071-0170 (Dec. 20, 2007), available at
Verne Kopytoff, Thwarting Microsoft Lifts Google, S.F. CHRON. (May 9, 2008) (reporting that Google CEO Eric Schmidt emphasized that “Google would not set the prices for ads on Yahoo! because those prices are determined by an auction.”).
See, e.g., Miguel Helft, The Humans Behind the Google Money Machine, N.Y. TIMES (June 2, 2008) (“Many [advertisers] say that despite efforts by Google to be more transparent, they remain in the dark about what goes on inside the company’s ad machine.”).
The long-term agreement was the culmination of coordination between the two companies over the last several months, including a two-week “test” that allowed Google and Yahoo! to obtain information about their relative search advertising pricing.
Indeed, Microsoft recently requested a copy of the agreement from both Google and Yahoo!. Neither company has responded
See, e.g., Yahoo!’s 8-K (June 12, 2008). Illustrations submitted to the SEC by Yahoo! suggest that the exact same search ad in the exact same location could be sourced by Google over Yahoo! under the deal. What is clear is that there is a limited amount of ad space on each Yahoo! search results page, and Google search ads will frequently displace Yahoo!’s in the most valuable positions. Any advertising space used by Google, of course, is space that an advertiser could no longer purchase directly from Yahoo!.
The agreement provides for the payment to Yahoo! of a percentage of Google’s revenues under the agreement, “with such percentage adjusting based on specified monthly gross revenues thresholds.” Thus, the agreement may be structured so that Yahoo! has the incentive to shift more traffic to Google to earn an even larger cut of Google’s growing revenues. In addition, according to the SEC filings, Google may terminate the agreement if certain revenue thresholds are not satisfied. Google’s termination right provides a strong incentive for Yahoo! to continue to rely on Google’s ads.
Yahoo! estimated in an SEC filing a few months ago that Google’s revenue per search is, on average, 60 to 70 percent higher than Yahoo!’s, which suggests that Yahoo! will want to cede control to Google for a fair number of Yahoo!’s advertisements. Exhibit 99.2, Yahoo! Investor Presentation at 14 (Mar. 2008) in Yahoo! Current Report (Form 8-K) (Mar. 18, 2008).
Ms. Decker has explained that Yahoo! will not generate any costs savings as a result of the deal. See Transcript of Yahoo!’s June 12, 2008 Conference Call to Discuss Advertising Agreement with Google at 5. Similarly, Google Chairman and CEO Eric Schmidt explained there will be little to no improvement to Google’s platform as a result of the deal. See Transcript of Google’s June 12, 2008 Conference Call to Discuss Advertising Agreement with Yahoo! at 6-7.
The antitrust laws treat such “naked” agreements among competitors less favorably than mergers or joint ventures that hold out the prospect for efficiencies, which could lead to the development of new technologies and products as well as cost savings.
Search advertising revenues reached $8.6 billion in 2007 and are expected to jump to almost $16.6 billion by 2011. That is comparable to both radio and cable advertising, whose revenues each totaled about $20 billion in 2007, but search advertising is growing much faster.
The Yahoo! document discussed whether Yahoo! would consider outsourcing search to Google and stated: “We are focused on long-term value creation rather than short term gains (short term analysis of the revenue potential of outsourcing monetization may not take into account the longer term impact on the competitive market if search becomes an effective monopoly).” See In re Yahoo! S’holders Litig., Cons. C.A. No. 3561-CC (Unsealed First Am. Verified Consol. Compl. ¶ 94, May 12, 2008), available at
Per se offenses are considered so egregious that courts are not permitted to assess the justification offered by the two parties entering into the agreement. The fact that the agreement exists, by itself, is enough to condemn the agreement as illegal.
See, e.g., APRIL 2000 DOJ/FTC ANTITRUST GUIDELINES FOR COLLABORATIONS AMONG COMPETITORS at § 3.34 (“The Agencies look to the competitive benefits and harms of the relevant agreement, not merely the formal terms of agreements among participants.”); id. at § 3.34(a) (“The Agencies inquire whether a collaboration is non-exclusive in fact as well as in name and consider any costs or other impediments to competing with the collaboration.”).
Digital marketing executives have voiced significant concern, such as Brian Waldman, a Boston-based Internet marketer (see Betsy Schiffman, Yahoo Customers Pan Google Ad Deal, WIRED EPICENTER BLOG (June 13, 2008), available at
), Arjo Ghosh, CEO iCrossing (see Sarah Arnott, Discontent flares over Google’s ‘dominance,’ THE INDEPENDENT (June 16, 2008), available at
), and Bryan Wiener, CEO 360i (see Abbey Klaassen, Google May Gain $1 Billion in Yahoo Ad Pact, Advertisers Suspect Prices Will Increase, ADVERTISING AGE (June 16, 2008)).