Attention rivalry among online platforms

Journal of Competition Law and Economics

Volumn 9, Issue 2, 2013, Pages 313-357

  Evans, David S ^a,b  

^a UNIVERSITY OF CHICAGO LAW SCHOOL   (United States)

^b UNIVERSITY COLLEGE LONDON   (United Kingdom)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 84878646508     PISSN: 17446414     EISSN: 17446422     Source Type: Journal    
DOI: 10.1093/joclec/nht014     Document Type: Article

References (87)

1. Online businesses for the purpose of this article include websites as well as web-based and native applications for mobile devices.
2. These online attention seekers may compete with offline attention seekers for attention and also for providing that attention to merchants and others who want access to it. This competition is increasingly important as the boundaries between online and offline attention seeking blur as they do, for example, in the increasing competition between online versus cable and over-the-air distribution of television shows and movies. This article focuses on examining competition among online attention seekers. It briefly discusses the implications for the broader analysis of competition among online and offline attention seekers but leaves the important topic of online and offline competition for future consideration.
3. Attention rivals are multi-sided platforms; they are intermediaries between suppliers and demanders of attention. For a survey of multi-sided platforms including advertising-supported ones
4. David S. Evans &Richard Schmalensee, The Antitrust Analysis of Multi-Sided Platform Businesses, in OXFORD HANDBOOK OF INTERNATIONAL ANTITRUST ECONOMICS (Roger D. Blair and D. Daniel Sokol eds., Oxford Univ. Press forthcoming 2013), available at http://papers.ssrn. com/sol3/papers.cfm?abstract_id=2185373.
5. For the purposes of this article, attention is the time that consumers spend focusing their minds on content. Attention rivals provide products and services to buyers, such as advertisers, who would like to get some of this attention. This attention is not necessarily fungible. Depending on the circumstances in which it is provided, advertisers might have a chance of getting more attention or greater focus from consumers or might get attention in a context that makes that attention more valuable because they have a greater chance of persuading consumers to buy something.
6. It is well known among economists that the demand substitution does not necessarily depend on products having similar functions and features. However, in practice, especially when competition authorities undertake the market definition exercise, it is common to focus on products that appear similar to each other. In any event, the novelty of this article is in arguing that the relevant dimension for assessing substitution is attention seeking and providing.
7. David S. Evans &Richard Schmalensee, Some Economic Aspects of Antitrust Analysis in Dynamically Competitive Industries, in 2 INNOVATION POLICY AND THE ECONOMY 1-50 (Adam B. Jaffe, Josh Lerner &Scott Stern eds., MIT Press 2002).
8. Douglas H. Ginsburg &Joshua D. Wright, Dynamic Analysis and the Limits of Antitrust Institutions, 78 ANTITRUST L.J. 1 (2012).
9. J. Gregory Sidak &David Teece, Dynamic Competition in Antitrust Law, 5 J. COMPETITION L. &ECON. 581 (2009).
10. For a survey of some of the arguments concerning "tipping to monopoly" in online network industries.
11. CARL SHAPIRO &HAL R. VARIAN, INFORMATION RULES: A STRATEGIC GUIDE TO THE NETWORK ECONOMY (Harv. Bus. Rev. Press 1998).
12. The argument presented in this article is also different from the well-known phenomenon that new attention-seeking technologies, such as television, substitute for older ones, such as radio. The difference involves the speed of introduction on the Internet and the fact that there is intense innovation competition to develop new attention-diverting properties.
13. This part incorporates key contributions by Catherine Tucker who pointed the author to the new channel attribution technologies discussed here. See Catherine Tucker, The Implications of Improved Attribution and Measurability for Online Advertising Markets (2012), available at http ://www.ipria.org/events/conf/Competition_Conference/Measureabiltiy_and_online_ads.pdf.
14. Guangdong Province High People's Court, Civil Judgment No. Yuegaofaminsanchuzi 2/2011, available at http://www.gdcourts.gov.cn/gdcourt/front/front!content.action?lmdm= LM43&gjid=20130328040159946185. I consulted with Tencent on this matter and a portion of the Court's analysis was based on the analytical framework described in this article. For further discussion of this case and an English-language translation of a portion of the decision, see David S. Evans, Vanessa Zhang &Howard Chang, Analyzing Competition Among Internet Players: Qihoo v. Tencent, CPI ANTITRUST CHRON.,May 2013.
15. Compete.com, Traffic and Engagement Metrics, Aug. 2012.
16. National Telecommunications and Information Administration, Households with Internet, 2001, http://www.ntia.doc.gov/files/ntia/publications/table_householdinternet2001.pdf (based on Bureau of the Census, Current Population Survey, September 2001 Internet and Computer Use Supplement, 2001); National Telecommunications and Information Administration, Household's Internet Connection Type, 2001, http://www.ntia.doc.gov/files/ntia/publications/table_ householdinternetconnection2001.pdf (based on Bureau of the Census, Current Population Survey, September 2001 Internet and Computer Use Supplement, 2011); National Telecommunications and Information Administration, Households Using the Internet In and Outside the Home, 2011, http://www.ntia.doc.gov/files/ntia/data/CPS2010Tables/t11_2.txt (based on Bureau of the Census, Current Population Survey, October 2011 School Enrollment and Internet Use Supplement, 2011).
17. National Telecommunications and Information Administration, Household's Internet Connection Type, 2001, http://www.ntia.doc.gov/files/ntia/publications/table_ householdinternetconnection2001.pdf (based on Bureau of the Census, Current Population Survey, September 2001 Internet and Computer Use Supplement, 2011)
18. National Telecommunications and Information Administration, Households Using the Internet In and Outside the Home, 2011, http://www.ntia.doc.gov/files/ntia/data/CPS2010Tables/t11_2.txt (based on Bureau of the Census, Current Population Survey, October 2011 School Enrollment and Internet Use Supplement, 2011).
19. Compete.com, Traffic and Engagement Metrics, Sept. 2002, Sept. 2007, Sept. 2012. Compete. com collects real-time clickstream data from users who have installed its toolbar, and combines this with clickstream data licensed from ISPs. The combined panel contains over two million U.S. Internet users. Normalization is used to make the panel representative of the population of U.S. users of web browsers. For more details, see Compete.com, Overview of Data Methodology and Practices, available at http://media.compete.com/site_media/upl/img/Compete_Data_Methodology_3.pdf. These data-and all of the other data on online use reported below-do not include mobile devices. Starting in the late 2000s, mobile devices-smart phones and tablets-that could access the Internet were available to consumers. Applications for these devices could reside on the web, in which case consumers access the application through their mobile phone browser, or the application could reside on the mobile device itself and provide both online and offline services
20. Fred Cavazza, Mobile Web App vs. Native App? It's Complicated, FORBES.COM, Sept. 27, 2011, http://www.forbes.com/sites/fredcavazza/2011/09/27/mobile-web-app-vs-native-app-its-complicated/.
21. See supra note 13 for a description of the several sources of data used for Table 1.
22. Not all of these sites are attention rivals that buy and sell attention. Some, such as chase.com and paypal.com, are merchant websites that get attention but do not resell it to other businesses. However, since these merchant websites have tended to stay in the top 50 for the entire decade, the evidence on exit from the top 50 would be even stronger if they had been excluded.
23. USC Annenberg School's Center for the Digital Future, Digital Future Report (2010, 2011). These figures do not include the use of mobile applications. The diversion of time from personal computers to mobile could explain a portion of the decline. However, it is unlikely that the use of mobile applications accounted for much of the drop during this time period. According to Nielsen data, 23 percent of mobile subscribers had smart phones in the first quarter of 2010. However, of this 23 percent, only 37 percent had Apple and Android phones, which were the only smart phones with significant application use. Therefore only 8.5 percent (37 percent of 23 percent) of mobile phone users had smart phones that could entail significant application use.
24. Based on an analysis of data from compete.com.
25. Netcraft, Netcraft Web Server Survey, Dec. 12, 2012, http://news.netcraft.com/archives/category/web-server-survey/.
26. Table 3 excludes websites that are not primarily attention-seekers, as well as cases where special circumstance can account for the changes in time spent at the site.
27. Over this period, some of the changes are the result of people shifting their attention from the use of PCs to mobile devices. For example, the shift in social networking likely reflects the use of Facebook using a PC to the use of Facebook from mobile devices.
28. While these data are suggestive, a more definitive analysis would require an analysis of diversion of traffic between sites and between categories.We lack the data to do that, but it should be one of the central exercises in competition policy matters to properly assess market definition.
29. Yahoo!, Annual Report 14 (Form 10-K) (Feb. 29, 2012).
30. Yahoo!, Annual Report (Form 10-K) (Feb. 25, 2004)
31. Yahoo!, Annual Report (Form 10-K) (Feb. 29, 2012).
32. Joanna Brenner, Pew Internet: Mobile, PEWINTERNET, Dec. 4, 2012, http://pewinternet.org/Commentary/2012/February/Pew-Internet-Mobile.aspx.
33. Times Topics: iPad and Tablet Computers, N.Y. TIMES, Oct. 23, 2012, http://topics.nytimes.com/top/reference/timestopics/subjects/i/ipad/index.html.
34. Zach Whittaker, comScore: Apple Continues to Gain on Google in U.S. Market Share, ZDNET, Nov. 2, 2012, http://www.zdnet.com/comscore-apple-continues-to-gain-on-google-in-u-smarket-share-7000006826/.
35. Jon Fingas, Google Play Hits 600,000 Apps, 20 Billion Total Installs, ENGADGET, June 27, 2012, http://www.engadget.com/2012/06/27/google-play-hits-600000-apps/.
36. Microsoft Unveils Windows Phone 8, MICROSOFT NEWS CENTER, Oct. 29, 2012, http://www. microsoft.com/en-us/news/press/2012/oct12/10-29windowsphone8pr.aspx.
37. The location-based application has a feature called "Find Lick." About Shopkick, SHOPKICK. COM, http://www.shopkick.com/about.
38. Jessica Vascellaro &Ian Sherr, Apple Veers Away From Google, WALL ST. J., June 12, 2012, http ://online.wsj.com/article/SB10001424052702303768104577460244284627170.html
39. Peter Kafka, Eric Schmidt's "Gang of Four" Doesn't Have Room for Microsoft, ALLTHINGSD.COM, May 31, 2011, http://allthingsd.com/20110531/eric-schmidts-gang-of-four-doesnt-have-room-formicrosoft/.
40. Network effects for many Internet sites are reversible, and the history of the Internet shows at least some examples of category leaders such as MySpace in social networking losing their dominant position in a category.
41. In many cases, these physical media charge consumers a price that is less than the marginal cost of production and distribution. Attention Rivalry Among Online Platforms 331
42. Neil J. Thurman &Jack Herbert, Paid Content Strategies for News Websites: An Empirical Study of British Newspapers' Online Business Models, 1 JOURNALISM PRACTICE 208 (2007).
43. Hany Nada, Music for Nothing and the Fans for Free, ALLTHINGSD.COM, Oct. 28, 2011, http ://allthingsd.com/20111028/music-for-nothing-and-the-fans-for-free/.
44. Chris Anderson, The Economics of Giving It Away, WALL ST. J., Jan. 31, 2009, online.wsj.com/article/SB123335678420235003.html.
45. Branden Sero, www.qq.com, SHVOONG.COM, June 20, 2012, http://www.shvoong.com/internetand-technologies/websites/2298647-www-qq-com/.
46. Jeff Bercovici, New York Post Pivots on iPad Strategy, Dropping Paywall, FORBES, June 19, 2012, http://www.forbes.com/sites/jeffbercovici/2012/06/19/new-york-post-pivots-on-ipad-strategydropping-paywall/.
47. Josh Halliday, Times Loses Almost 90% of Online Readership, GUARDIAN, July 20, 2010, http ://www.guardian.co.uk/media/2010/jul/20/times-paywall-readership.
48. Mitra Pathak, Zuckerberg: Facebook to Intensify Competition with Google with More Robust Search Engine, TOPNEWS, Sept. 12, 2012, http://www.topnews.in/zuckerberg-facebook-intensifycompetition-google-more-robust-search-engine-2365005.
49. Nick Zieminiski &Bikoy Koyitty, Insight: Facebook, LinkedIn Threaten to Slay Monster.com, REUTERS, Aug. 24, 2011, http://uk.reuters.com/article/2011/08/24/us-monster-socialmediaidUKTRE77N6QW20110824.
50. Careers at Facebook, FACEBOOK.COM, Dec. 17, 2012, https://www.facebook.com/careers/teams/engineering.
51. Some of these new websites do not represent new businesses, but are rather new websites for existing businesses.
52. Database, Compete.com, Jan. 2012, July 2012.
53. Compete.com, Sept. 2002 to Sept. 2012. exceptions involve the fact that comScore aggregates different domains owned by the same firm.
54. Under the narrower definition, a website counts as an advertising-supported attention seeker if (1) it is consumer-focused, and (2) its revenues come exclusively or almost exclusively from banner ads, search ads, or flat listing fees for items for sale. Under the broader definition, a website counts as an attention-seeker if (1) it is consumer focused, (2) it earns substantial revenue from banner ads, search ads, fees for items for sales (whether flat listing fees, commissions, revenue sharing, buyer fees, or sales lead fees), (3) its other revenue comes from advertising-related sources, such as consumer research, marketing campaign design, other marketing services, premium membership fees for consumers to avoid ads or receive additional services, and credits for virtual goods.
55. LinkedIn's products and services include the following: LinkedIn Connections, Invitations, People You May Know, Addressbook Importer, LinkedIn Today, Personalization Platform, LinkedIn Alumni, Mobile, APIs, Widgets, LinkedIn Corporate, Solutions/Recruiter, LinkedIn Referral Engine, LinkedIn Recruitment Media, Job Seeker Basic, Job Seeker Plus, LinkedIn Ads, Display Ads, Custom Groups, Recommendation Ads, InMail, Profile Organizer, and LinkedIn for Salesforce. LinkedIn, Inc., Annual Report 4 (Form 10-K) (Mar. 2, 2012).
56. See LinkedIn Developers, Why Develop with Us, https://developer.linkedin.com/whydevelop.
57. Mikolaj Piskorski, LinkedIn (A) (Harvard Business School Case Study, 9-707-406, Feb. 26, 2007), available at static.ow.ly.docs/LinkedIn%20(A)_ouN.pdf.
58. Search engines are similar to printed yellow pages. The providers of these business directories sell advertising on the directory pages.
59. Tucker, supra note 9.
60. According to marketing material for Visual IQ's IQ Intelligence Suite, which is one of the leading cross-channel attribution technologies, "Once your marketing performance data and customer data (and any other data you wish to include) has been collected, formatted, normalized and integrated during the software set-up process, IQ Insight presents it to you through its powerful, yet easy to use dashboard and reporting interface. This allows you the flexibility to view, report upon and compare the performance of every channel, campaign and marketing tactic side by side, and to analyze that performance by any criteria specific to your company, industry or business model. IQ Insight provides both executive overview and granular views of your marketing ecosystem using a common set of your own key performance indicators (KPIs)." Intelligence Suite, VISUALIQ.COM, http://www.visualiq.com/products/iqintelligence-suite (last visited Nov. 9, 2012).
61. Econsultancy (2012). The data were collected from September 26 to October 23, 2011; of 607 respondents, 44 percent were located in North America and 33 percent were in the United Kingdom.
62. Forrester Interactive Attribution (2012).
63. For a survey, see Evans and Schmalensee, The Antitrust Analysis of Multi-Sided Platform Businesses, supra note 3.
64. Susan Athey, Emilio Calvano &Joshua S. Gans, The Impact of the Internet on Advertising Markets for News Media (Rotman School of Management, Working Paper No. 218051, Nov. 22, 2012), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2180851. These authors develop a model that assumes that some people use one platform while others stochastically search the web. They show that under this realistic assumption many of the results of earlier models that assumed single-homing are reversed.
65. Mark Armstrong, Competition in Two-Sided Markets, 37 RAND J. ECON. 668 (2006).
66. Different sources of attention may have different values to advertisers because of differences in "how attentive" people are to advertisements on the property and the context in which they are seeing the advertisements. These differences between sources of attention lead to differences in the likelihood that a view by a consumer will result in a "conversion"-that is, a sale of a good- for the merchant. However, these differences do not necessarily lead to traditional product differentiation issues. For example, if property A leads to 10 percent of the conversions per person viewing an advertisement as property B, and that is the only difference, the price for showing an ad to a person on property B would be 10 percent of the price of the same advertisement on property B. At these prices the two properties would be close substitutes.
67. For example, it is much "cheaper" per click to buy display ads on Facebook than to buy search ads on Google. But the Facebook ads result in much fewer conversions per click than do the Google ads. Larry Kim, Google Display Network vs. Facebook Advertising, WORDSTREAM, May 15, 2012, http://www.wordstream.com/blog/ws/2012/05/15/ipo-facebook-vs-google-displayadvertising; JCD Repair, Google Adwords vs. Facebook Ads-It Was No Contest, June 19, 2012, http://www.jcdrepair.com/blog/google-adwords-vs-facebook-ads-it-was-no-contest/645.
68. JCD Repair, Google Adwords vs. Facebook Ads-It Was No Contest, June 19, 2012, http://www.jcdrepair.com/blog/google-adwords-vs-facebook-ads-it-was-no-contest/645
69. Avi Goldfarb &Catherine Tucker, Substitution Between Offline and Online Advertising Markets (Working Paper, Nov. 30, 2010), available at http://ssrn.com/abstract=1721001.
70. Chantal Tode, Macy's Makes Mobile Integral Part of Black Friday Strategy to Drive In-Store Sales, MOBILE COMMERCE DAILY, Nov. 6, 2012, http://www.mobilecommercedaily.com/macy%E2% 80%99s-enhances-mobile-app-to-facilitate-shopping-on-black-friday
71. Chantal Tode, Fairmont Hotels Enhances Foursquare Check-Ins with Location-Based Offers, MOBILE COMMERCE DAILY, Aug. 31, 2012, http://www.mobilecommercedaily.com/fairmont-hotels-enhances-foursquarecheck-ins-with-location-based-offers.
72. Skype and Microsoft provide communication services to consumers and enterprises. For the sake of brevity, this part just considers the merger issues relevant for the consumer services.
73. Commission Decision, Case No. Comp/M.6281-Microsoft/Skype, 2011 O.J. (C 341) 2, 20, available at http://ec.europa.eu/competition/mergers/cases/decisions/m6281_20111007_20310_ 2079398_EN.pdf.
74. Press Release, Monster Worldwide, TMP Worldwide Reaffirms Commitment to HotJobs Acquisition (Dec. 12, 2001), available at http://www.about-monster.com/content/tmpworldwide-reaffirms-commitment-hotjobs-acquisition. At that time, Monster was owned and operated by TMP, a global human resources firm, which started in 1967.
75. John Christie, Jr. &Wendy Anderson Terry, Federal Trade Commission Hints at Tough Position on Mergers Between Online Competitors, WILMERHALE, Mar. 21, 2002, http://www.wilmerhale.com/pages/publicationsandNewsDetail.aspx?NewsPubId=92779.
76. For example, the popular HR software system, PeopleSoft Talent Management, has a tool that helps assess the ROI of different recruiting methods. "Oracle's talent management modules cover all phases of the talent lifecycle-planning, recruiting, performance, learning, career development, succession planning, compensation, talent reviews and measuring and reporting." Peoplesoft Talent Management, ORACLE, http://www.oracle.com/us/products/applications/peoplesoft-enterprise/human-capital-management/talent-management/features/index.html.
77. Justin Pope, Online Job Sites Face Changed Landscape; Lack of Jobs: Monster and Hotjobs See Earnings Decline, ASSOCIATED PRESS, Feb. 24, 2002, http://www.seacoastonline.com/apps/pbcs. dll/article?AID=/20020204/BUSINESS/302049981.
78. Robin Goldwyn Blumenthal, Turning into a Monster of a Competitor, BARRON'S, Sept. 21, 2009, available at http://online.barrons.com/article/SB125332142936624651.html#articleTabs_ article%3D1.
79. There were reports in mid-2012 that Facebook had plans to start a job board as well. Joseph Walker, Facebook to Launch Job Board, WALL ST. J. ONLINE, July 6, 2012, http://online.wsj.com/article/SBB0001424052702304141204577510933875917766.html.
80. Compete.com, Sept. 2002.
81. The author worked for Monster on this matter and made presentations to the FTC. This part, however, is based solely on public information.
82. See Guangdong Province High People's Court, Civil Judgment No. Yuegaofaminsanchuzi 2/2011. As mentioned earlier, I consulted Tencent in this matter and developed material that Tencent introduced in its presentation to the Court and that became the basis for a significant part of the Court's analysis.
83. iResearch data on number of users per month (Feb. 2012).
84. TENCENT HOLDINGS LIMITED, 2011 ANNUAL REPORT 107 (2011), available at http://www. tencent.com/en-us/content/ir/rp/2011/attachments/201102.pdf.
85. Qihoo 360 Technology Co. Ltd, Registration Statement (Form F-1) (Mar. 14, 2011), available at http://www.sec.gov/Archives/edgar/data/1508913/000104746911002124/a2202432zf-1.htm.
86. iResearch data on number of users per month (Feb. 2012); these are proprietary data and not publicly available.
87. Li Bin, Multisectoral Come Forward to "Call a Timeout" Intermission "3Q War", BEIJING TIMES REPORTER, Nov. 5, 2010, http://epaper.jinghua.cn/html/2010-11/05/content_ 600180.htm.