The Federal Trade Commission concluded its investigation into Google purchase of AdMob on May 21, 2010, after an extensive review in which the FTC sought to assess competitive effects in a particularly challenging context.
1. challenge facing the FTC was that the mobile advertising industry in which these firms participated was and is a rapidly changing, nascent industry. AdMob, which the FTC described as the leading firm in the industry, had been in existence for only four years when the FTC completed its review of the merger.
2. The difficulty inherent in identifying and weighing potential anti competitive effects in a very fluid environment was evidently one the FTC was not shy about confronting.
Another challenge, less evident at the beginning of the review process but certainly more salient now, is that the competition in mobile advertising is linked to competition in other markets. The acquisition of another mobile advertising firm by Apple near the end of the FTC’s merger review provided clear evidence that competition to monetize software loaded on smart phones was part of a larger competition between mobile operating systems (“mobile OS” vendors such as Apple, Google, and Microsoft.
In this article, we discuss some of the economic issues that arise in undertaking a competitive effects analysis in technologically dynamic industries such as mobile advertising.
II. Merger Review in Rapidly Evolving Industries
Mergers in rapidly changing industries pose a challenge to the antitrust agencies and the parties because there is often substantial uncertainty about the path the industry will take even in the near term. The usual steps in merger analysis – defining a relevant market, assessing potential unilateral and coordinated effects, and determining the likelihood of entry – may require confronting substantially more uncertainty in a fluid, innovation-driven industry. At the same time, there are large and potentially durable welfare effects from failing to prevent mergers that would reduce the incentive to innovate. High rates of innovation can both make the effects of the merger more difficult to assess and make predicting the effects correctly particularly important.
Before turning to the more difficult issues presented in the Google-AdMob merger, we first turn to the simple context of the Thoratec-Heartware merger to explore how merger analysis has responded to the need to account for both the effect of innovation on the expected path of industry evolution and the effect of innovation on competition.
Rapid innovation does not always make the application of standard competitive effects analysis difficult. Sometimes the likely path of competition that an industry will follow can be foreseen with some confidence even though the industry is experiencing innovation and change. The proposed Thoratec- Heart Ware merger, which was abandoned by the parties after the FTC announced its intent to challenge the transaction, is a case that has been cited by the FTC as an example of how to apply the competitive effects analysis called for in the updated Horizontal Merger Guidelines.3 In 2009,
Thoratec Corp. supplied the only FDA approved left ventricular assist device, a mechanical heart assist device implanted in patients with end-stage heart disease. While supplying the currently dominant device, Thoratec was also developing a next generation device. Heart Ware International, Inc., a small research firm with no FDA-approved products, was also developing a next generation product in competition with Thoratec and several other medical device companies. In early 2009, the companies announced the planned acquisition of Heart Ware by Thoratec Corp.
4The FTC review focused on competition involving next generation devices and the effect the acquisition might have on the prices as well as any potential reduction in the pace of innovation.
This required the FTC to forecast whether Heart Ware was likely to be an important supplier of next generation devices. Despite rapid innovation in these heart assist devices, which might have made this kind of forecast highly speculative, the regulatory requirements for approval of medical devices together with the FTC’s extensive interviews of investigators utilizing the devices in clinical trials provided the FTC with some guidance as to the path the industry would likely follow. The FTC concluded that, of the firms developing next generation products, “only Heart Ware poses a potential threat [to Thoriate].”
5 The FTC also concluded that if and when the FDA approved Heatwave new device, commercial supply of Heart Ware’s product could be expected to “rapidly erode Thoriate monopoly.”6 The FTC reasoned that the merged firm would anticipate the cannibalization of its profits from the introduction of the Heart Ware device and would therefore have an incentive to delay its introduction. If, however, Heart Ware remained an independent firm, it would have no monopoly profits to erode and would therefore push to get its product on the market as soon as possible. Given the available evidence on the probable path of innovation from the device approval process and the FTC’s field investigation, the FTC decided it could adequately predict the path competition would most likely take and filed a complaint to block the merger.
Although the analysis in this case was aided by information that may not be available in all instances, it is important to note that the FTC’s reasoning depended on a nuanced view of the likely innovation process. For the FTC’s logic to be correct, it is not enough that it correctly foresee that Heart Ware would win the innovation competition by producing a superior device.
It must also be the case that the hypothesized delay imposed by the merged firm in the introduction of HeartWare’s winning design would not allow a third competitor to introduce an effective competing device during the period of delay. If the competition were closely contested, the merged firm could not preserve its profits by delaying introduction of the new device. Although there appear to have been several other companies pursuing innovative treatments for end-stage heart failure, how this would have played out has the merger proceeded is impossible to know. At the time of writing, Heart Ware’s device has still not received FDA approval.
The information available to the FTC for assessing future competition in the mobile advertising industry was quite different in the case of the Google AdMob merger. There was no regulatory process to provide information on the probable path of innovation or on which firms would be future participants in mobile advertising. Nor was there a well-defined trajectory of innovation that suggests successive product generations as one sees, for example, in the microprocessor industry. While this heightened uncertainty was acknowledged by the FTC in its investigation of the Google AdMob transaction,7 the FTC nonetheless undertook an extensive merger review and, despite closing its investigation, apparently concluded that the deal had the potential to have substantial anti competitive effects.8 According to reports, the FTC had been poised to block the merger,9 only to close its investigation after Apple entered the mobile advertising industry by acquiring an AdMob competitor.
To understand the difficulty facing the Commission in its review of this merger, it is helpful to have some background on the industry. The small but rapidly growing mobile advertising industry exists to monetize applications and other software on mobile devices. Mobile ads are advertisements that appear on smartphones and on tablet computers. The ads appear in mobile search results, on websites specifically designed for mobile devices, and in applications installed on mobile devices (“apps”).
The birth and growth of this industry is the result of the introduction and proliferation of internet-capable mobile phones (generically, “smartphones”). Smartphones have captured a steadily increasing share of U.S. mobile phone sales since Apple introduced the iPhone in 2007, currently comprising 37.4% of mobile phone sales and expected to comprise 50% of mobile phone sales by September 2012.10 The expanding use of smartphones has, in turn, triggered rapid growth in the mobile advertising industry, from $368 million of revenue in 2009 to $877 million in 2010.
11 At the time of the merger AdMob was a small startup that provided a service (called “ad serving”) that matched mobile apps that offered spaces in which advertisements could appear with advertisers that had advertisements to be placed in those spaces. The advertiser would pay the ad server for placing the ad and the server would, in turn, pass a share of that payment on to the app developer. The ad serving industry was crowded, rapidly growing, and highly fluid when the proposed acquisition was announced in late 2009. At that time, AdMob faced over 18 competitors in the United States, including AOL, Google, Jumptap, Microsoft, Millenial, Nokia, Quattro, and Yahoo.12
A number of factors made it difficult for the FTC to assess the relative strength of these competitors and the likely competitive effects of the merger. Market share information was sparse and market participants questioned the validity of the few public estimates available.13 Further, information on differentiating product characteristics and substitution in demand between firms was scarce. Determining which of the many participants in the industry were close competitors was further complicated by the ease with which the participating firms could and did change business models, a fluidity that is a common characteristic of nascent industries. Existing firms could change their business models in response to the evolution of demand and quickly reposition themselves to be closer competitors than they had been. For example, InMobi was focused on mobile web ads in 2009, not ads within apps, but was serving ads to both mobile webpages and apps in 2010.14 Finally, entry into the industry was ongoing and continued throughout the FTC’s review process: Google and AdMob faced at least 22 competitors by the time the merger was approved, some of which were significant.15 InMobi, for example, had served markets outside the
U.S. and had developed the largest mobile advertising network in Asia.16 As of 2011, InMobi was the second largest mobile advertising network worldwide.17 In combination, these factors introduced substantial uncertainty into any prediction as to the likely path of competition in the mobile advertising industry.
To understand the nature of competition in the industry and the direction it might take, the FTC sought extensive third party discovery.18 But despite the FTC’s best efforts, the industry was evolving too rapidly for its review process to keep pace. Market participants who met with the FTC on multiple occasions noted that the industry changed so quickly that some information they had provided to the FTC became stale from one meeting to the next.19 Other industry participants approached by the FTC felt that the industry was too fluid for anyone, even people intimately familiar with the technology and the competing firms, to confidently predict the effects of the merger.20 Indeed, the entry by Apple that led FTC knows enough to support a decision to block the deal. . . . More generally, it seems obvious
the FTC to reverse course exemplifies the difficulty of conducting a merger review in a rapidly changing industry.
Nevertheless, after examining the facts, the FTC determined that Google and AdMob were the two leading firms in the industry or – and the public statements are somewhat ambiguous on this point – in a segment of the industry described as “performance ads.”21 Mobile ads typically allow the user to select them to go to a site on the internet, performing the mobile equivalent of a “click through.” Performance ads charge advertisers by the number of click-throughs. The other prevalent pricing model in the industry is to charge (usually a lower price) by the number of times the ad is displayed to a mobile device user. Based on its understanding of competition in the industry, the FTC appeared to be willing to challenge the merger.
III. Mergers and Platform Competition
The growth of the mobile device market has led to fierce competition among the suppliers and adopters of the Android, Apple, and Microsoft mobile operating systems in a battle that has been named the “smartphone war.” This competition, which is reflected in patent acquisition and litigation as well as in product innovation, has affected a broad range of industries and firms, including the device manufacturers supplying smartphones and other mobile devices, the software developers writing applications for these devices, and the participants in the mobile advertising industry. The effects on competition in the mobile advertising industry were acknowledged in the FTC’s public statements about its Google-AdMob decision:
“AdMob also competes with Google in the sale of mobile advertising on Google’s Android platform. Android and iPhone compete against each other as platforms, and the availability of free of low-cost applications helps drive that competition.”28
The economic logic underlying these statements seems to invoke a view of systems competition that goes beyond the standard competitive effects analysis of conduct within a single industry. By recognizing that the competition in mobile advertising can affect competition in mobile platforms, the FTC is extending competitive effects analysis to take account of competitive effects in markets other than the one in which the merger occurs. While this is not new territory, it makes for a more complex analysis. In this instance, the agency’s reasoning appears to recognize that how Google competes in mobile advertising will affect and be affected by competition in mobile operating systems.
Google’s business model is to make money from internet and mobile advertising. To be successful in the mobile segment, Google needs access to mobile devices and to information about those mobile devices.29 Targeting restaurant ads effectively, for example, requires knowing where the user is located. Even more fundamentally, formatting an ad correctly requires knowing the kind of device on which it will be displayed. The access to mobile device information necessary to serve ads effectively can be compromised by contractual restraints placed on application developers by the mobile OS/platform owner. As evidenced by Apple’s iOS app developer contracts, the platform owner can limit the access to crucial information that facilitates ad service by its competitors.
One way Google can defend against this strategy is by ensuring that it has access to the large share of mobile devices that are based on its Android platform. Broad deployment of the Android operating system, however, depends on Google’s ability to persuade OEMs to adopt it. One powerful inducement to adopt the Android operating system is its price; the Android operating system is offered free of charge by Google.30 Another is to ensure that there will be many apps available for Android-based devices so that end-users will choose to buy Android devices.