Federal Communications Commission DA 14-1862 Before the Federal Communications Commission Washington, D.C. 20554 In the Matter of Implementation of Section 6002(b) of the Omnibus Budget Reconciliation Act of 1993 Annual Report and Analysis of Competitive Market Conditions With Respect to Mobile Wireless, Including Commercial Mobile Services ) ) ) ) ) ) ) ) ) WT Docket No. 13-135 SEVENTEENTH REPORT Adopted: December 18, 2014 Released: December 18, 2014 By the Chief, Wireless Telecommunications Bureau: TABLE OF CONTENTS Heading Paragraph # I. INTRODUCTION .................................................................................................................................. 1 II. COMPETITIVE DYNAMICS WITHIN THE INDUSTRY ................................................................ 10 A. Service Providers ............................................................................................................................ 11 1. Facilities-Based Providers ....................................................................................................... 11 2. Resale and MVNO Providers ................................................................................................... 15 3. Other Providers ........................................................................................................................ 17 B. Connections, Net Additions, Churn ................................................................................................ 19 1. Subscribers and Total Connections, and Net Additions .......................................................... 19 2. Subscribers and Net Additions ................................................................................................. 23 3. Churn ....................................................................................................................................... 26 C. Market Shares and Concentration .................................................................................................. 29 D. Financial Indicators ........................................................................................................................ 34 1. Revenue and ARPU ................................................................................................................. 34 2. Average Revenue per Unit (ARPU) by Provider ..................................................................... 38 3. Wireless Telephone Services CPI ............................................................................................ 40 4. Profitability Metrics ................................................................................................................. 42 III. OVERALL MOBILE WIRELESS INDUSTRY METRICS ................................................................ 44 A. Network Coverage .......................................................................................................................... 45 1. Overall Network Coverage ...................................................................................................... 46 2. Network Coverage and Roaming ............................................................................................. 57 3. Network Coverage by Technology .......................................................................................... 58 4. Network Coverage by Income Levels ...................................................................................... 60 5. Commission Actions Related to Coverage, Technology and Roaming ................................... 61 B. Connections and Subscribers .......................................................................................................... 66 1. Connections and Subscribers by Geography ........................................................................... 66 1 Federal Communications Commission DA 14-1862 2. Connections and Subscribers by Demographics ...................................................................... 67 C. Consumers and Mobile Wireless .................................................................................................... 69 1. Usage ....................................................................................................................................... 70 2. Handsets ................................................................................................................................... 75 3. Mobile Applications................................................................................................................. 77 4. Consumer Access to Information ............................................................................................. 82 5. Intermodal Developments ........................................................................................................ 86 IV. INPUT MARKETS ............................................................................................................................... 87 A. Spectrum ......................................................................................................................................... 88 1. Importance of Spectrum for the Provision of Mobile Wireless Services ................................. 89 2. Facilitating Access to Spectrum ............................................................................................... 93 3. Analysis of Spectrum Holdings ............................................................................................. 104 B. Non-Spectrum Input Segments..................................................................................................... 108 1. Wireless Infrastructure ........................................................................................................... 108 2. Backhaul ................................................................................................................................ 116 C. Downstream Mobile Wireless Ecosystem .................................................................................... 120 V. PRICING LEVELS AND TRENDS................................................................................................... 126 A. Postpaid Service ........................................................................................................................... 128 1. Equipment Installment Plans ................................................................................................. 129 2. Changes in Monthly Pricing of Postpaid Plans ...................................................................... 135 3. Impact of EIPs on Consumer Costs ....................................................................................... 140 4. Promotions and Incentives ..................................................................................................... 144 5. Evolution of Data Pricing Models ......................................................................................... 147 6. Other Developments .............................................................................................................. 152 B. Prepaid Service ............................................................................................................................. 154 1. Prepaid Plan Choices ............................................................................................................. 159 2. Availability of Handsets ........................................................................................................ 162 3. Service Coverage ................................................................................................................... 163 4. Data Speeds and Data Allowances ........................................................................................ 164 C. Price Indicators for Mobile Data .................................................................................................. 165 1. Postpaid Smartphone Data Price ............................................................................................ 166 2. Prepaid Smartphone Data Price ............................................................................................. 167 VI. NON-PRICE RIVALRY .................................................................................................................... 168 A. Investment .................................................................................................................................... 169 B. Network Coverage and Technology Upgrades ............................................................................. 173 1. Current Coverage by Provider ............................................................................................... 174 2. Coverage and Roaming .......................................................................................................... 180 3. Service Provider Network Deployments................................................................................ 182 C. Quality of Service ......................................................................................................................... 189 1. Network Speed ....................................................................................................................... 193 2. Latency ................................................................................................................................... 204 D. Differentiation in Mobile Wireless Handsets/ Devices ................................................................ 208 E. Advertising and Marketing ........................................................................................................... 212 VII.CONCLUSION ................................................................................................................................... 216 VIII. PROCEDURAL METTERS ........................................................................................................ 217 APPENDICES APPENDIX I: Maps APPENDIX II: Competitive Dynamics within the Industry APPENDIX III: Overall Mobile Wireless Industry Metrics APPENDIX IV: Input Markets 2 Federal Communications Commission DA 14-1862 APPENDIX V: Pricing Levels and Trends APPENDIX VI: Non-Price Rivalry APPENDIX VII: List of Abbreviations and Acronyms 3 Federal Communications Commission DA 14-1862 I. INTRODUCTION 1. In this Seventeenth Mobile Wireless Competition Report (Report), the Federal Communications Commission (Commission or FCC) fulfills its obligation, pursuant to Section 332(c)(1)(C) of the Communications Act, to report annually to the Congress on the state of competition in mobile services. Competition in mobile wireless services is a cornerstone of the Commission’s mission and essential for driving innovation, investment, and consumer benefits. In recent years, mobile wireless services have gone from a luxury to a convenience to an absolutely central part of Americans’ daily lives. Increasing numbers of users now have multiple devices connected to mobile networks. Handsets are no longer used just for voice communication, email, social networking, and web browsing, but increasingly as hubs for entertainment, mobile commerce, and to connect other personal devices such as smart watches and fitness monitors. These developments have helped make mobile wireless one of the most important sectors in the national economy. 2. Following on the Sixteenth Report, released in March 2013, which provided an analysis of market conditions and developments during 2010, 2011, and 2012,1 this Report presents data and analysis covering 2013 and the first half of 2014, to the extent data are available.2 The analysis focuses on “competitive market conditions with respect to commercial mobile services,” as required by the Act.3 While like the Sixteenth Report, this Report presents a multitude of industry data on various aspects of mobile wireless competition,4 it employs a more data-centric model, with a more concise analysis along with a greater use of Tables and Charts in accessible data formats. For instance, we are providing the charts and tables in the Report and its Appendices, as well as much of the underlying data, on a dedicated website5 that we intend to update before the release of the next Report as new data becomes available. 3. Similar to previous reports, the analysis in this Report is based on a consumer-oriented view of mobile services, with a focus on specific product categories regardless of their regulatory classification. Thus, our analysis of commercial mobile radio services (CMRS) is integrated into an analysis of all mobile wireless services, including not only voice, but also messaging and broadband.6 Because consumers increasingly view various mobile voice, messaging, and data services as interchangeable with one another, no matter their regulatory classification, service providers are competing for customers using CMRS services as well as non-CMRS services. As a result, the Commission has indicated that it is important to consider potential substitutes when analyzing the competitive landscape for these services, and to evaluate the mobile wireless industry as a whole, rather than just focusing on the provision of CMRS services.7 This Report analyzes competition across the entire 1 Implementation of Section 6002(b) of the Omnibus Budget Reconciliation Act of 1993, Annual Report and Analysis of Competitive Market Conditions with Respect to Commercial Mobile Services, Sixteenth Report, 28 FCC Rcd at 3836 (2013) (Sixteenth Report). 2 For instance, much data are only published as year-end numbers and are publicly available only in middle of the following year. For example, all CTIA data are yearend 2013, and this is available in their annual report published in July 2014. For these data, we are able to present only 2013 numbers with no mid-year updates. As more data becomes available, we plan to provide web updates rather than wait until release of the next Competition Report. 3 47 U.S.C. § 332(c)(1)(C),. 4 47 U.S.C. § 332(c)(1)(C). As with previous Reports, this Report does not address the merits of any license transfer applications that are currently pending before the Commission or that may be filed in the future, which will be decided based on the record collected in each proceeding. 5 fcc.gov/wireless-competition-report 6 See Section II, Introduction, infra. 7 See Sixteenth Report at 28 FCC Rcd at 3836 (2013) at ¶31. As the Commission has concluded, paraphrasing the Department of Justice/Federal Trade Commission guidelines on merger review, “When one product is a reasonable substitute for the other in the eyes of consumers, it is to be included in the relevant product market even though the products themselves are not identical.” Application of Echostar Communications Corporation, General Motors Corporation, and Hughes 4 Federal Communications Commission DA 14-1862 mobile wireless marketplace, including key market segments such as spectrum and infrastructure. This Report, like the previous three Reports, adopts an approach similar to the earliest reports, but undertakes an expanded and more detailed competitive analysis of the entire mobile wireless ecosystem. 4. Congress enacted the requirement in 1993 that the Commission report annually on “competitive market conditions with respect to commercial mobile services.”8 At the same time, it created the statutory classification of “commercial mobile services” to promote the consistent regulation of mobile radio services that are similar in nature,9 and established the promotion of competition as a fundamental goal for CMRS policy formation and regulation.10 In particular, the statute requiring the annual report on CMRS competition states: The Commission shall review competitive market conditions with respect to commercial mobile services and shall include in its annual report an analysis of those conditions. Such analysis shall include an identification of the number of competitors in various commercial mobile services, an analysis of whether or not there is effective competition, an analysis of whether any of such competitors have a dominant share of the market for such services, and a statement of whether additional providers or classes of providers in those services would be likely to enhance competition.11 5. This Report complies with the statutory requirements for analyzing competitive market conditions with respect to commercial mobile services by employing an analysis founded upon an expanded view of the mobile wireless services marketplace and an examination of competition across the entire mobile wireless ecosystem. We analyze competitive rivalry in the mobile wireless industry, and the benefits received by consumers. This competitive analysis also identifies areas where competition is strong, as well as areas that could benefit from increased competition. 6. Consistent with the Commission’s first seven Reports, and the Fourteenth and subsequent Reports, this Seventeenth Report does not reach an overall conclusion or formal finding regarding whether or not the CMRS marketplace was effectively competitive, but provides an analysis and description of the CMRS industry’s competitive metrics and trends. 12 Given the complexity of the various inter-related segments and services within the mobile wireless ecosystem, we refrain from providing any single conclusion because such an assessment would be incomplete and possibly misleading in light of the variations and complexities we observe. Rather, the Report focuses on presenting the best data available on competition throughout this sector of the economy, both at the regional and national level, and highlighting several key trends in the mobile wireless industry. We note that there is no definition of “effective competition” widely accepted by economists or Electronics Corporation (Transferors) and Echostar Communications Corporation (Transferee), Hearing Designation Order, 17 FCC Rcd 20559, 20606 ¶ 106 (2002). 8 47 U.S.C. § 332(c)(1)(C). As noted in previous Reports, any individual proceeding in which the Commission defines relevant product and geographic markets, such as an application for approval of a license transfer, may present facts pointing to narrower or broader markets than any used, suggested, or implied in this Report. See, e.g., Implementation of Section 6002(b) of the Omnibus Budget Reconciliation Act of 1993, Annual Report and Analysis of Competitive Market Conditions with Respect to Commercial Mobile Services, Twelfth Report, 23 FCC Rcd 2241, 2250 ¶ 3 n. 5 (2008) (Twelfth Report). 9 Omnibus Budget Reconciliation Act of 1993, Pub. L. No. 103-66, Title VI, § 6002(b), amending the Communications Act of 1934 and codified at 47 U.S.C. § 332(c). 10 47 U.S.C. § 332 (a)(3). 11 47 U.S.C. § 332 (c)(1)(C). 12 This is in contrast to the Eighth through the Thirteenth Reports, which included a specific finding that there was effective competition in the CMRS market without defining the term “effective competition.” See, e.g., Thirteenth Report, 24 FCC Rcd 6185. 5 Federal Communications Commission DA 14-1862 competition policy authorities such as the U.S. Department of Justice (DOJ).13 The approach taken in this Report is consistent with the policy of the DOJ. 7. This Report first provides an analysis of the overall competitive dynamics of the industry, describing the various types of entities and their positions vis a vis one another across indices such as market share and various financial indicators.14 The Report then presents a broad overview of trends and developments in the mobile marketplace that have taken place since the Sixteenth Report, such as subscribership growth, adoption and deployment of technologies, and usage trends. While most of the developments have been along a continuum of previously noted trends, the ongoing deployment and adoption of LTE networks and the technologies they have enabled, has had a particularly profound effect throughout the mobile wireless marketplace during the period under review. 8. The Report then turns to an analysis of key inputs necessary for provision of mobile service, such as spectrum resources and network infrastructure. Spectrum, in particular, is the single most important input that wireless providers need for the provision of service and is a finite and scarce resource. The Report examines how the distribution of spectrum in the various bands affects competition. The Report next examines developments in the ways providers compete for and attract subscribers through pricing innovations, such as the decreased reliance on traditional handset subsidies and term contracts. As part of this analysis, the analysis looks at the differences between pre and postpaid market segments as well as ways in which those segments are converging. Finally, the Report analyzes competitive rivalry in non-price factors, such as coverage, service quality and speed of providers’ service offerings. 9. In addition to providing analysis of market conditions, various sections of the Report highlight Commission policies and actions designed to enhance competition -- for example, by making more spectrum available to existing mobile service providers and potential new entrants through competitive bidding such as the upcoming AWS-3 and incentive auctions. We also revised our transaction review process and spectrum screen to ensure that multiple providers in each market have access to sufficient spectrum to compete effectively.15 The Commission’s policies have been guided by the goal of promoting and preserving competition, which in turn has facilitated the ability of consumers to make choices among numerous service providers and leads to lower prices, improved quality, and increased innovation.16 II. COMPETITIVE DYNAMICS WITHIN THE INDUSTRY 10. As part of our analysis of competition in the mobile wireless services industry, we begin by discussing some of the various competitive dynamics within the industry.17 Providers of mobile wireless services 13 See Ex Parte Submission of the United States Department of Justice, GN Docket No. 09-51 at 11 (filed Jan. 4, 2010). The DOJ states, “[t]he operative question in competition policy is whether there are policy levers that can be used to produce superior outcomes, not whether the market resembles the textbook model of perfect competition.” 14 . Dollar figures stated in this Report have not been adjusted for inflation (i.e., they are nominal dollars) unless stated otherwise. 15 See In The Matter Of Policies Regarding Mobile Spectrum Holdings Expanding The Economic And Innovation Opportunities Of Spectrum Through Incentive Auctions, WT Docket No. 12-269, Report and Order (Mobile Spectrum Holdings Report and Order), 29 FCC Rcd 6133 16 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6144¶ 17, and at 6193¶ 143. Our public interest evaluation necessarily encompasses the “broad aims of the Communications Act,” which include, among other things, a deeply rooted preference for preserving and enhancing competition in relevant markets, accelerating private sector deployment of advanced services, and generally managing the spectrum in the public interest. See, e.g., AT&T WCS Order, 27 FCC Rcd at 16464 ¶ 11; AT&T-Centennial Order, 24 FCC Rcd at 13928 ¶ 28;. 17 We discuss in more detail in Sections IV and V below additional aspects in the competitive dynamics of the industry when we discuss elements of price and non-price rivalry. 6 Federal Communications Commission DA 14-1862 offer an array of mobile voice and data services, including interconnected mobile voice services, text and multimedia messaging, and mobile broadband Internet access services. Mobile wireless services also include machine-to-machine connections for fleet management systems, smart grid devices, vehicle tracking, home security systems, and other telematics services. This section presents information and data on all mobile wireless services as well as on individual services and segments where appropriate and when the data are available.18 A. Service Providers 1. Facilities-Based Providers 11. Facilities-based mobile wireless service providers offer mobile voice, messaging, and/or data services primarily using their own network facilities, although coverage areas usually are supplemented through roaming agreements.19 Facilities-based providers can operate nationwide, multi-regional, regional, or local networks. Some data and messaging services offered by facilities-based providers rely only on Internet Protocol (IP)-based, packet-switched networks, but most mobile voice services continue to connect to the Public Switched Telephone Network (PSTN) and rely on North American Numbering Plan telephone numbers. 12. Nationwide Service Providers. As of year-end 2013, there were four facilities-based mobile wireless service providers in the United States that industry observers typically describe as “nationwide.” These providers include AT&T,20 Sprint,21 T-Mobile,22 and Verizon Wireless.23 Although none of these four providers has a network that covers the entire land area or population of the United States, each has a network that covers a significant portion of both, and therefore these four providers will be referred to as “nationwide providers” throughout this Report.24 Each of the four nationwide service providers has a mobile wireless network that covers in excess of 99 percent of the U.S. population.25 18 See Sixteenth Report, 28 FCC Rcd at 3836 at ¶ 20 - 22 19 Fixed wireless services are currently not included in our analysis of mobile wireless services. 20 AT&T Mobility is the successor of Cingular, a joint venture formed in October 2000 between Southwestern Bell (SWB) and Bell South. In 2005, SWB subsidiary that included SWB’s interest in Cingular merged with AT&T Corp. In 2006, AT&T acquired Bell South and therefore 100% ownership of Cingular. As of December 31, 2013, AT&T Mobility served more than 110 million subscribers.” See AT&T Inc., SEC Form 10-K, filed Feb. 21, 2014, at 2. 21 Sprint Nextel was created by the merger in 2005 of Sprint Corp. and Nextel Communications, Inc. See Tenth Report, 20 FCC Rcd at 15931 ¶ 60. On July 5, 2013, the FCC released an order approving the acquisition of Sprint by SoftBank Corp., and Sprint’s acquisition of 100 percent of Clearwire’s stock. See Softbank-Sprint-Clearwires Order, 28 FCC Rcd at 9643-44 ¶ 1-4. 22 T-Mobile traces its roots to May 2001,when Deutsche Telekom AG (DT) acquired in a deal worth $24 billion two US providers -- VoiceStream Wireless (formerly a division of Western Wireless that had recently acquired regional GSM providers Aerial Communications in the Midwest and Omnipoint in the Northeast), and Southern regional provider Powertel. In September 2002, they were re-branded nationally with the T-Mobile name, conforming to the brand under which DT provided mobile services overseas. See http://www.celtnet.org.uk/telecos/T-mobile.php Most recently, on March 12, 2013, the FCC released an order approving the application of Deutsche Telekom, T-Mobile, and MetroPCS, which resulted in the creation of T-Mobile USA as a wholly-owned subsidiary of Deutsche Telekom . See T-Mobile-MetroPCS Order, 28 FCC Rcd at 2323-24 ¶ 1-2. 23 As of December 31, 2013, Verizon owned a controlling 55% interest in Verizon Wireless and Vodafone owned the remaining 45%. See Verizon Communications, Inc., SEC Form 10-K, filed Feb. 27, 2014, at 2. On February 21, 2014, Verizon completed its acquisition of Vodafone’s 45 percent indirect interest in Verizon Wireless, so that Verizon now owns 100 percent of Verizon Wireless. See http://newscenter.verizon.com/corporate/news-articles/2014/02-21-acquisition-of- vodafone-stake-in-vzw-complete/ 24 All four nationwide have spectrum in CONUS and in HI and AK. 25 Thus, a nationwide network covers a sufficiently large percentage of the population such that it would be inappropriate to categorize it as a regional network. 7 Federal Communications Commission DA 14-1862 13. All nationwide providers provide service directly to consumers and businesses and also provide machine-to-machine (M2M) services. Later in the Report, detailed data and analysis are provided on the retail voice and broadband service provided by these companies. However, there are limited statistics on M2M communications. The research firm “Current Analysis” estimates that AT&T had approximately 14.7 million M2M connections, Verizon Wireless had between 7 and 9 million connections, and both Sprint and T-Mobile had 3.3 million connections. M2M has gained significant interest in the past few years as providers continue to provide connectivity between devices, sensors, monitors, etc. and their networks. The new “Internet of Things” (IoT) is seen by some commentators as promising the next major opportunity for providing interconnection and advanced connect among devices. Many industries such as healthcare, are transforming to use M2M networks to connect their numerous smart devices and machines. While M2M resides mostly in the Enterprises space, more and more providers are launching services for the home market. 14. Multi-Regional, Regional, and Local Service Providers. US Cellular is a multi-regional service provider that has developed wireless networks and customer service operations covering five geographic market areas in portions of 23 states that collectively represent a total population of 31.8 million as of December 31, 2013.”26 US Cellular relies on roaming agreements with nationwide facilities-based providers, as well as other smaller providers, to supply service to its customers in areas not covered by its networks. C-Spire and Ntelos are two other regional providers with substantial market presence in certain parts of the country. There are also dozens of regional and local facilities-based providers throughout the continental United States, Alaska, and Hawaii that typically provide service in a single geographical area, many of them rural areas.27 2. Resale and MVNO Providers 15. Resellers and mobile virtual network operators (MVNOs) do not own any network facilities but instead purchase mobile wireless services wholesale from facilities-based providers and resell these services to consumers.28 An agreement between an MVNO and a facilities-based provider may be more likely to occur when the MVNO has better access to some market segments than the host facilities-based provider, possibly due to its brand reputation, distribution network, marketing strategies, or business model.29 MVNOs often increase the range of services offered by the host facilities-based provider by targeting specific market segments, including segments previously not served by the hosting facilities-based provider.30 Hence, the relationship between an MVNO and its hosting facilities-based provider can be a mutually beneficial strategic partnership.31 In 2013, the 26 United States Cellular Corp., SEC Form 10-K for the fiscal year ended December 31, 2013 at 1. 27 Some regional facilities based providers include, but are not limited to, Alaska Communications, Big River Broadband, Bluegrass Cellular, Cellcom, Choice Wireless, GCI Wireless, People’s Wireless, Pioneer, West Central Wireless. 28 According to one service provider, “MVNOs execute a contract with [the facilities-based provider] to buy wireless service from [the facilities-based provider] to resell under their own brand to customers and perform all marketing, billing, collections and customer service for the customers they activate. MVNOs establish and maintain the relationship with its customers. MVNOs own the relationship with their customers and establish their own calling plans and pricing.” See Verizon Wireless, Authorized Retailers and MVNOs, http://www.verizonwireless.com/b2c/aboutUs/reseller/authorizedAgentIndex.jsp (visited June. 23, 2014). 29 See P. Kalmus and L. Wiethaus, On the Competitive Effects of Mobile Virtual Network Operators, Telecommunications Policy, Vol. 34, 2010 at 263, 266, 268. 30 See P. Kalmus and L. Wiethaus, On the Competitive Effects of Mobile Virtual Network Operators, Telecommunications Policy, Vol. 34, 2010, at 268 (On the Competitive Effects of Mobile Virtual Network Operators). See A. Banerjee and C. Dippon, Voluntary Relationships Among Mobile Network Operators and Mobile Virtual Network Operators: An Economic Explanation, Information Economics and Policy, Vol. 21, 2009, at 72 (Voluntary Relationships Among Mobile Network Operators and Mobile Virtual Network Operators: An Economic Explanation). 31 See The Yankee Group, Jason Armitage, Yankee Group’s 2011 Predictions: 4G Fuels the Decade of Disruption, at 7 (stating, “[I]t’s critical the MVNO does not compete to any meaningful degree with the host.”) 8 Federal Communications Commission DA 14-1862 largest MVNO was TracFone Wireless (TracFone).32 16. Unlike facilities-based providers, MVNOs do not engage in non-price rivalry by creating capacity through network investments, network upgrades, or network coverage. MVNOs may target their service and product offerings at specific demographic, lifestyle, and market niches, including consumers who are low income, are relatively price sensitive, do not want to commit to multi-year subscription contracts, have low usage needs, or do not want to buy a bundle that contains unwanted data services. Following widespread industry practices, the Commission generally attributes the subscribers of MVNOs to their host facilities-based providers, including when it calculates market concentration metrics. 3. Other Providers 17. Narrowband Data Providers. Narrowband data and paging services comprise a specialized market segment of the mobile wireless industry. These services include two-way messaging, as well as machine- to-machine and other telemetry communications, and are consumed primarily by businesses, government users, and other institutions. According to XXXX licensing databases, there is approximately seven megahertz of spectrum allocated to narrowband and paging services, and there are hundreds of licensees for these services, including private individuals, firms, and local and state governments. 18. Mobile Satellite Service Providers. Mobile Satellite Services (MSS) providers offer satellite- based communications to mobile devices. Traditionally, MSS has involved voice and narrowband data services. MSS services are generally targeted at users who require service in remote areas, in disaster response situations, or other places where terrestrial mobile wireless network access may be limited.33 Examples of MSS customers include the oil industry, maritime users, public safety agencies, and other government/military operations. B. Connections, Net Additions, Churn 1. Subscribers and Total Connections, and Net Additions 19. In the period since the Sixteenth Report, the U.S. mobile wireless services industry experienced continued strong growth, with total wireless connections up by 10 million in 2013.34 Of the four nationwide facilities-based providers, AT&T and Verizon Wireless maintained the largest market shares throughout 2013.35 T-Mobile had the largest quarterly increases in market share during this time period. While Sprint steadily lost subscribers in the first three quarters of 2013, it rebounded slightly in the final quarter of the year.36 20. This Report uses several data sources to estimate the number of mobile wireless subscribers and connections. One source, the Numbering Resource Utilization Forecast (NRUF), tracks the quantity of phone numbers that have been assigned to mobile wireless devices.37 Based on NRUF data, it appears that the number 32 Some MVNO companies that currently provide service include Straight Talk, H2O Wireless, Ultra Mobile, Net10, LycaMobile, Spot Mobile, Telcel America, GIV Mobile, Simple Mobile, Red Pocket, Pure Talk, PagePlus, Ting, iWireless, Voyager, FreedomPop, ROK Mobile, Tracfone, See sprint/?utm_source=GeneralUsers&utm_campaign=41e3559dd9- c:tec,mdad:07-29&utm_medium=email&utm_term=0_1dd83065c6-41e3559dd9-98996217 http://gigaom.com/2014/07/28/unlock-phone-att-verizon-tmobile 33 See Thirteenth Report, 24 FCC Rcd at 6301 ¶ 247. 34 NRUF, Dec 2013. 35 UBS Investment Research. US Wireless 411 v51 4Q13. Figure 21 36 UBS Investment Research. US Wireless 411 v51 4Q13, Figure 24 37 When all mobile wireless devices were assigned telephone numbers and subscribers generally carried one mobile device for making voice calls, NRUF provided reasonably accurate measures of subscribership. Now, however, consumers are more likely to use more than one mobile device that have been assigned telephone numbers – particularly non-voice devices, such as Internet access devices (e.g., wireless modem cards, netbooks, and mobile Wi-Fi hotspots), e-readers, tablets, and telematics systems. In addition, certain mobile broadband providers do not assign telephone numbers to at least some of the devices on their networks. Therefore, NRUF is becoming less useful in measuring the number of individual subscribers. 9 Federal Communications Commission DA 14-1862 of mobile wireless connections in 2013 were 335.7 million. Connections grew three percent during 2013, from 329.2 million at the end of 2012, to 339.2 million at the end of 2013. CTIA also estimated the total number of mobile wireless connections based on its own industry survey,38 and found that the number of connections grew by three percent during the same period, from 326.5 million at the end of 2012, to 335.7 million at the end of 2013.39 This information is presented in Chart II.B.1. Note. Based on data from NRUF, CTIA (CTIA Year-End 2013 Wireless Indices Report, Table 6). Latest available data. 21. Chart II.B.2 presents data on total connections by service segment. The postpaid segment accounts for more than 60 percent of the total connections over the reported period, while the prepaid connections have grown from approximately 15 to 20 percent of the total connections. Wholesale connections and connected devices are a small but growing portion of the total connections.40 Instead, it is providing more of an estimate of the number of mobile wireless connections or connected devices. In addition, it will become a less accurate measure of connected devices to the extent that more devices are sold that do not use telephone numbers. 38 CTIA states that “the terms subscriber, subscriptions, and connections are being used interchangeably” in their report and survey. See CTIA Wireless Industry Indices at p. 7. 39 See Appendix Table II.B.i for detailed data. 40 See Appendix Table II.B.ii for detailed data. 0 50 100 150 200 250 300 350 400 E st im at ed C on ne ct io ns ( in m il li on s) Year Chart II.B.1 Total Mobile Wireless Connections 2001 - 2013 NRUF CTIA 10 Federal Communications Commission DA 14-1862 Source: UBS Investment Research. US Wireless 411 Version 51. Figure 17: US Wireless 411 Version 54. 22. Table II.B.1 presents data on total connections of the larger individual providers Based on the 2014 data, it appears that AT&T and Verizon Wireless account for roughly two-thirds of the estimated connections, with Sprint and T-Mobile together accounting for slightly less than a third. Regional providers accounted for approximately 3 percent of the total mobile wireless connections. Table II.B.1 Estimated Total Connections for Publicly Traded Facilities–Based Mobile Wireless Service Providers (In thousands) 2011- 1st Half 2014 Nationwide Providers 2011 2012 2013 1st Half 2014 Verizon Wireless 108,667 116,570 125,535 129, 615 A&T 103,247 106,965 110,276 116,542 Sprint 55,021 55,626 54,622 54,080 T-Mobile 30,756 30,299 46,684 50,545 Nationwide Provider Total 297,691 309,460 337,117 350,782 Regional Providers 2011 2012 2013 2nd Q 2014 US Cellular 5,891 5,798 4,774 4,653 Metro PCS 9,347 8,887 Leap Wireless 5,934 5,297 4,551 NTELOS 415 440 465 458 Cincinnati Bell 459 398 340 277 Regional Provider Total 22,046 20,820 10,130 5,388 Total Estimated Connections 319,736 330,279 347,247 356,170 Note: UBS Wireless 411 Report. Version 51 2014 Q1, Table 21, pp.14. UBS Wireless 411 Version 54. Total 0 50,000 100,000 150,000 200,000 250,000 300,000 350,000 400,000 1Q11 2Q11 3Q11 4Q11 1Q12 2Q12 3Q12 4Q12 1Q13 2Q13 3Q13 4Q13 1Q14 2Q14 T ot al C on ne ct io ns (t ho us an ds ) Quarter, Year Chart II.B.2 Qualrterly Total Mobile Wireless Connections by Service Segment 2011-1st Half 2014 Postpaid Prepaid Wholesale Connected devices 11 Federal Communications Commission DA 14-1862 estimated connections figure includes data only for the providers reported in the table. Annual numbers are end of the year, 1st half 2014 numbers are end of 2nd quarter 2014. 2. Subscribers and Net Additions 23. Net additions for 2013 totaled 10.0 million based on NRUF data, and 9.2 million based on CTIA data. The Commission is also able to use data reported by service providers on Form 477 to estimate the number of mobile voice subscribers and mobile internet subscribers. Form 477 data generally show a lower number of subscriber additions than NRUF data. Between 2011 and 2013, annual net subscriber additions amounted to 12.2 million, 7.6 million, and 4.2 million, respectively. This information, along with NRUF and CTIA data, is presented in Chart II.B.3. Source: NRUF, CTIA (CTIA Year-End 2013 Wireless Indices Report Table 6), Form 477. Latest available data. 24. In 2012, quarterly net additions varied by service segment, with connected device adds seeing significant growth. In the fourth quarter of 2012, wholesale additions dropped substantially, while postpaid adds showed significant increase. The net number of connected device additions was consistently higher than prepaid additions during through the first half of 2014, with the second quarter of 2013 and 2014 showing negative prepaid additions. Postpaid net additions fell dramatically in the first quarter of 2013, but climbed for the remainder of the year and showed significant growth in the first half of 2014. This information is presented in Chart II.B.4.41 41 See Appendix Table II.B.iii for detailed data. 16.6 11.1 11.1 15.5 11.9 10.0 14.9 15.3 10.7 19.7 10.5 9.2 12.0 13.0 10.8 12.2 7.6 4.2 0.0 5.0 10.0 15.0 20.0 25.0 2008 2009 2010 2011 2012 2013 N et A dd it io ns ( in M ill io ns ) Year Chart II.B.3 Annual Net Additions to Total Mobile Wireless Connections (2008 - 2013) NRUF CTIA Form 477 12 Federal Communications Commission DA 14-1862 Source: UBS Investment Research. UBS Wireless 411 Version 54. Figure 16: UBS categorizes Tracfone customers as prepaid, not wholesale. 25. From 2009 through 2011, AT&T had the largest number of net additions, partly through acquisitions and partly through organic growth. Verizon Wireless had the largest number in 2012, accumulating the most net additions in 2012 and 2013. T-Mobile hovered between third and last place from 2009 to 2012. In 2013, however, T-Mobile experienced a surge in customer growth, moving past AT&T to accumulate the second most net additions for the year, as well as in the first half of 2014. The significant subscriber growth Sprint experienced during 2010 and 2011 slowed significantly in 2012, and Sprint lost customers in 2013. US Cellular lost subscribers every year from 2009 through the first half of 2014. These trends are displayed in Chart II.B.5. 1Q12 2Q12 3Q12 4Q12 1Q13 2Q13 3Q13 4Q13 1Q14 2Q14 Total 3,533 2,246 2,156 3,682 2,181 788 3,229 5,389 4,378 4,418 Prepaid 1,891 414 462 603 1,278 -1,391 280 1,069 472 -1,029 Wholesale 1,296 568 1,244 -151 3,431 598 436 802 -945 975 Connected Devices 493 480 854 1,053 1,344 1,318 1,381 1,026 1,703 1,574 Postpaid -147 784 -405 2,177 -3,872 263 1,132 2,492 3,147 2,899 3,533 2,246 2,156 3,682 2,181 788 3,229 5,389 4,378 4,418 -1000 0 1000 2000 3000 4000 5000 6000 7000 8000 N et A dd s (t ho us an ds ) Chart II.B.4 Quarterly Net Additions by Service segment 2012 - 1st Half 2014 13 Federal Communications Commission DA 14-1862 Source: UBS Investment Research. UBS Wireless 411 Version 51. Figure 14: UBS Wireless 411 Version 54. 3. Churn 26. Churn measures the number of connections that are disconnected from mobile wireless service during a given period time period, and is usually expressed as a percentage. Churn is calculated by dividing the aggregate number of wireless subscriber connections who canceled service during a period by the total number of wireless subscriber connections at the beginning of that period. The churn rate for the period is equal to the average of the churn rate for each month of that period, e.g., the three months in a quarter or the twelve months for an annual churn rate. Thus a monthly churn rate of 1 percent averaged over the three month reporting period would also be reported as 1 percent. Providers publish their monthly churn rate information as part of their quarterly filings with the SEC. 27. A service provider’s churn rate depends on many factors, including the distribution of its customers between postpaid and prepaid service plans, customer satisfaction with their service provider, service provider switching costs, and competition. As an example, if a service provider has an average monthly churn rate of 2 percent in each month of a year, the service provider would lose approximately 24 percent of its customer base over the course of the year. 28. Churn rates of the nationwide facilities-based service providers, measured in the second quarter of 2014, ranged from 1.2 percent for Verizon Wireless and 1.5 percent for AT&T, up to 2.1 percent for T-Mobile and 2.4 percent for Sprint. These data are presented in Chart II.B.6. The average industry quarterly churn rates have ranged from 1.9 percent to 3.0 percent since 2010.42 Churn rates for prepaid connections are typically significantly higher than churn rates for postpaid connections, because prepaid customers, unconstrained by a 42 UBS Investment Research. US Wireless 411 v51 Q1 2014. Figure 14: U.S. Wireless Industry Model. 2009 2010 2011 2012 2013 1st Half 2014 Verizon 5,486 5,627 6,421 7,903 8,868 4220 AT&T 7,278 8,853 7,699 3,764 2,721 1696 Sprint -1,141 1,777 5,111 605 -2,500 -796 T-Mobile 1,032 -56 -549 204 4,300 3861 US Cellular -55 -69 -186 -88 -1,024 -121 -4,000 -2,000 0 2,000 4,000 6,000 8,000 10,000 N et S ub sc ri be r A dd it io ns ( T ho us an ds ) Chart II.B.5 Annual Net Additions by Service Provider (2009-1st Half 2014) 14 Federal Communications Commission DA 14-1862 multi-month or multi-year service contract, are more likely than postpaid customers to terminate a relationship with a wireless service provider. Source: UBS Investment Research. US Wireless 411 Version 49, Table 16; US Wireless 411 Version 51, Figure 28; US Wireless 411 Version 54. C. Market Shares and Concentration 29. Revenues and connections or subscribers are key metrics that are used to measure the size of a company. In turn, the relative size of a company compared to the total size of the industry determines market share. The revenue data are presented in Table II.C.1 below. Table II.C.1 Service Revenues for Facilities–Based Mobile Wireless Service Providers (In millions of dollars) 2005-1st Half 2014 National 2005 2006 2007 2008 2009 2010 2011 2012 2013 1H2014 Verizon 28,131 32,796 38,016 49,717 52,046 55,629 59,157 63,733 69,033 36,065 AT&T 30,665 33,788 38,678 44,249 48,563 53,510 56,726 59,186 61,552 30,535 Sprint 28,631 31,918 32,106 28,435 25,832 25,894 27,390 29,086 29,263 14,337 T-Mobile 12,308 14,511 16,891 19,242 18,926 18,689 18,481 17,213 20,535 10,821 Nextel 1,695 468 Regional 2005 2006 2007 2008 2009 2010 2011 2012 2013 1H2014 US Cellular 2,832 2,445 3,679 3,940 3,926 3,913 4,054 4,099 3,595 1,697 Metro PCS 872 1,291 1,919 2,437 3,130 3,690 4,428 4,540 Leap (Cricket) 769 956 1,396 1,709 2,171 2,413 2,829 2,947 2,631 NTELOS 264 302 357 392 400 383 395 424 467 226 Cincinnati Bell 215 236 267 291 284 269 252 225 185 80 Alltel 6,485 7,030 7,984 Centennial 395 433 484 524 408 CentennialPCS 363 339 294 320 236 Dobson 1,117 1,202 1,030 1Q1 1 2Q1 1 3Q1 1 4Q1 1 1Q1 2 2Q1 2 3Q1 2 4Q1 2 1Q1 3 2Q1 3 3Q1 3 4Q1 3 1Q1 4 2Q1 4 AT&T 1.4% 1.4% 1.3% 1.4% 1.5% 1.2% 1.3% 1.4% 1.4% 1.4% 1.3% 1.4% 1.4% 1.5% Verizon Wireless 1.3% 1.2% 1.2% 1.2% 1.2% 1.1% 1.2% 1.2% 1.3% 1.2% 1.3% 1.2% 1.3% 1.2% Sprint 2.3% 2.2% 2.3% 2.2% 2.2% 2.0% 2.1% 2.3% 2.4% 3.2% 2.3% 2.1% 2.4% 2.4% T-Mobile 3.4% 3.3% 3.5% 4.0% 3.3% 3.2% 3.4% 3.7% 3.3% 2.6% 2.5% 2.5% 2.1% 2.1% 0.0% 0.5% 1.0% 1.5% 2.0% 2.5% 3.0% 3.5% 4.0% 4.5% C hu rn R at e Chart II.B.6 Quarterly Churn Rate for Nationwide Mobile Wireless Providers 2011 - 1st Half 2014 15 Federal Communications Commission DA 14-1862 Rural Cellular 511 539 608 327 SunCom 739 755 649 Note: UBS Investment Research. - UBS Wireless 411 Report. Version 51, Table 31. UBS Wireless 411 Report Version 54, MetroPCS was acquired by T-Mobile in March 2013 and Leap (Cricket) was acquired by AT&T in March of 2014. 30. Market share is usually calculated as the percentage of an industry or market's total revenues earned (or number of customers served) by a particular company over a specified time period.43 In general, market share increases and decreases can be a sign of the relative competitiveness of a company's products or services. Nationwide service provider market shares by service revenues are shown in Table II.C.2 below. This table provides market share estimates of the largest facilities-based service providers based on revenues. The four nationwide service providers accounted for about 96 percent of the nation’s mobile wireless service revenue in 2013, up from 91.5 percent in 2012. The service revenues of Verizon Wireless and AT&T accounted for about 70 percent of total service revenue in 2013. Table II.C.2 Market Shares for Facilities-Based Mobile Wireless Providers Based on Service Revenues 2011 – 201344 Nationwide Service Providers 2011 2012 2013 Verizon Wireless 33.8% 34.4% 36.5% AT&T 32.4% 32.0% 32.5% Sprint 15.6% 15.7% 15.5% T-Mobile 10.6% 9.3% 10.9% Total National Service Provider Market Share 92.4% 91.5% 95.3% Regional Service Providers 2011 2012 2013 US Cellular 2.3% 2.2% 1.9% Metro PCS 2.5% 2.5% Leap Wireless 1.6% 1.6% 1.4% NTELOS 0.2% 0.2% 0.2% Cincinnati Bell 0.1% 0.1% 0.1% Other 0.7% 1.9% 1.0% Total Regional Service Provider Market Share 7.6% 8.5% 4.7% Note: Data based on Table II.C.1, infra ,UBS Wireless 411 Report. Version 51 2014 Q1, Table 31, pp.19 UBS Wireless 411 Report. Version 54 and CTIA total service revenue figures. For 2011, the data are also from the Sixteenth Competition Report Table 11 and 12. 31. Market concentration can be measured by the number of competitors in the marketplace, or by the share of subscribers, sales or revenues attributable to each competitor. High market concentration levels in a given market may raise some concern that the market is not competitive. However, an analysis of other factors, such as prices, entry conditions, and non-price rivalry, may find that a market with high concentration levels is competitive. 32. The Herfindahl-Hirschman Index (HHI), which is employed by the Commission to measure 43 Markets as discussed in this Report are independent of markets determined in the context of transactions. In prior transactions, the Commission has found that the relevant geographic markets for certain wireless transactions generally are “local” and have used CMAs (cellular marketing areas) as the local geographic market. In addition, it has also evaluated a transaction’s competitive effects at the national level where a transaction exhibits certain national characteristics that provide cause for concern. See AT&T-Leap Order, 29 FCC Rcd at 2735 ¶ 27. 44 We do not report mid-year market shares. 16 Federal Communications Commission DA 14-1862 market concentration, is a widely-accepted measure of concentration in competition analysis. The HHI is calculated by summing the squared market shares of all firms in any given market. In this Report, we calculate HHIs by EA (economic areas), to maintain continuity with past Reports and to ensure that we do not compromise the confidential information found in the NRUF data.45 The Commission generally estimates HHIs on a narrower geographic area than the EA to evaluate the competitive consequences of transactions.46 33. At the end of 2013, the weighted average of the HHI (weighted by population across the 172 Economic Areas in the United States) for the mobile wireless services industry was 3,027, a small increase from 2,966 at the end of 2012, which in turn was an increase from 2,874 at the end of 2011.47 As in previous years, the most recent increases in the weighted average of HHIs reflect continued industry consolidation, such as the 2013 merger of T-Mobile and MetroPCS. Average HHIs across EAs are presented in Chart II.C.148 At the end of 2013, the value of the HHI for individual Economic Areas (EAs) ranged from a low of 2,237 in EA 63 (Milwaukee-Racine WI) to a high of 6,689 in EA 146 (Missoula MT). Source: NRUF and 2010 census data, EAs defined as in 1995. The latest NRUF data available is 2013. 45 By contrast, in wireless transactions, the Commission has analyzed competitive effects on “local” geographic markets using CMAs (cellular marketing areas). See AT&T-Leap Order, 29 FCC Rcd at 2735 ¶ 27. See also note 43 supra. 46 Antitrust authorities in the United States generally classify markets into three types: Unconcentrated (HHI < 1500), Moderately Concentrated (1500 < HHI < 2500), and Highly Concentrated (HHI > 2500). See Horizontal Merger Guidelines, U.S. Department of Justice and the Federal Trade Commission, http://www.justice.gov/atr/public/guidelines/hmg-2010.pdf . The Commission’s HHI screen flags markets for further competitive review if the HHI is 2800 with a change from the pre to the post transaction HHI of 100 or greater or a change of 250 or greater regardless of the initial HHI. See Applications of AT&T Wireless Services, Inc., Transferor, and Cingular Wireless Corp., Transferee, Memorandum Opinion and Order, 19 FCC Rcd 21522 (2004); Applications of AT&T Inc. and Cellco Partnership d/b/a Verizon Wireless for Consent to Assign or Transfer Control of Licenses and Authorizations and Modify a Spectrum Leasing Arrangement, WT Docket No. 09-104, Memorandum Opinion and Order, 25 FCC Rcd 8704 (2010). 47 Data are based on 2010 census data. EAs defined as in 1995. 48 EA level data are presented in Appendix Table II.C.i 2,693 2,806 2,866 2,874 2,966 3,027 - 500 1,000 1,500 2,000 2,500 3,000 3,500 2008 2009 2010 2011 2012 2013 H H I Year Chart II.C.1 Average HHI Across EAs 2008-2013 Highly Concentrated (HHI > 2500) 17 Federal Communications Commission DA 14-1862 D. Financial Indicators 1. Revenue and ARPU 34. Total service revenues of the wireless providers, as reported by the CTIA, include monthly service fees, usage-related charges, activation charges, vertical services (voice mail, enhanced calling features, and other services), out-collect roaming revenues, and data service revenues.49 In 2013, total wireless service revenue was $189 billion. It has grown steadily over the last twenty years, although the pace has tended to fluctuate over the last five years. Revenue increased by 2.3 percent in 2013.50 As reported earlier in Table II.C.1, company-specific revenues reported by the providers during this time indicate that Verizon Wireless and AT&T had the highest service revenues, followed by Sprint, and then T-Mobile.51 35. Given the longstanding variation in terms of plan characteristics and pricing for mobile voice and data, average price metrics have been necessary and useful tools to compare broad trends in pricing, even though average metrics have always had their limitations.52 Previous Reports have reported average price metrics, including the per-minute price of voice service, the average revenue per text message, the average revenue per megabyte, and the average revenue per unit (ARPU).53 As detailed in the Sixteenth Report,54 however, in the second half of 2012 Verizon and AT&T launched their shared data plans that bundled unlimited voice and texting with a data allowance for a single flat monthly fee, starting a shift toward shared data plans by wireless service providers. Beginning with its report of 2012 data, CTIA discontinued separately tracking and reporting wireless data service revenues.55 As a result of these changes, estimates of the unit price of wireless voice and data revenues are increasingly unreliable and difficult to come by, and the Commission is no longer able to report from the CTIA data an average revenue per text message, an average revenue per megabyte, or an average voice revenue per minute. 36. Similarly, the ARPU metric, which has commonly been used in the industry as an overall pricing indicator and which remains the best such measure currently used by industry and financial analysts, has come under increasing pressure, especially as a measure for comparisons across different providers. Consistent estimation of ARPU has become more difficult due to the growing prevalence of shared data plans, family and group plans, and other types of bundling, along with the increasing number of accounts with multiple devices, with each device potentially subject to a different pricing model. We note that ARPU is not consistently reported by different providers. For instance, AT&T, Sprint and T-Mobile report ARPU, while Verizon Wireless provides its reporting based on ARPA (Average Revenue per Account), which accounts for the multiple devices that may be associated to a single account. CTIA’s Wireless Industry Indices no longer report an “average local 49 See CTIA Year-End 2013 Wireless Indices Report, at 70-72. 50 Revenue increased by 3.0 percent, 4.9 percent, 6.2 percent, 9.0 percent, and 2.3 percent in 2009, 2010, 2011, 2012, and 2013, respectively. See Appendix Table II.D.i for details. 51 Detailed service provider service revenues are shown in Appendix Table II.C.i. 52 Different mobile wireless providers have offered a variety of pricing plans for their voice and data services, with service often offered under multi-part pricing schemes and with differing non-price terms and features, such as early termination fees and the consequences of reaching usage limits. As discussed in previous reports, it is therefore difficult to identify sources of information that track actual mobile wireless service prices in a comprehensive and consistent manner. Additionally, data on subscribership is not available at the plan level, and any average price comparison implicitly assumes uniform subscribership of all plans. 53 Industry and financial analysts have used company reported ARPU as a fair proxy of the amount of revenue generated per subscriber on a monthly basis 54 Sixteenth Report, 28 FCC Rcd at 3836, ¶144-146 55 See CTIA Year-End 2013 Wireless Indices Report, at 85-90 for a detailed discussion. 18 Federal Communications Commission DA 14-1862 monthly bill” consistently reported by providers that was used to estimate ARPU.56 To account for these shortcomings, industry and financial analysts have had to make additional assumptions and begun to estimate a new, normalized version of ARPU, dividing overall reported service revenues by the average number of subscribers for the period. 37. In its year-end 2013 annual report, CTIA reported an industry average measure of ARPU, specifically “Average Revenue per Reported (subscriber) Unit”,57 which is based “upon total revenues divided by the average total reported active units per survey period, divided by the number of months in the survey period,” i.e., an annualized monthly ARPU. The total service revenues used in this ARPU calculation includes roaming revenues, usage fees, access and other connection fees. Thus all revenues, including those from roamers in a provider’s market, are attributed to the subscriber base of the provider.58 According to CTIA, from December 2012 to December 2013, the average revenue per active unit (ARPU) in 2013 was $48.79 based on annual revenues and average active revenue-generating subscriber units. Total wireless industry service revenues for 2013 equaled $189.12 billion, up from $185 billion for 2012 as a whole. Total reported prepaid revenues for 2013 equaled $22.4 billion, down 5 percent from $23.7 billion reported for 2012.59 Chart II.D.1 below shows the total service revenue, subscribers and ARPU for the past 20 years. It appears that based on nominal dollars, the average industry ARPU appears to be fairly stable, while revenues and subscribership has increased. Note: Based on CTIA Year-End 2013 Wireless Indices Report, Table 27. Latest available data. 2. Average Revenue per Unit (ARPU) by Provider 38. There is considerable variation in ARPU amongst the various national and regional wireless 56 CTIA’s Indices Report provides discussion of some of the alternative methods of calculating ARPU. For example, the report indicates that a Yankee Group study on ARPU had found that the majority of major providers used nine to ten separate revenue components in their calculation. As noted in the Indices Report, consistency of reporting is critical to accurately compare the data over time. CTIA Year End 2013 Wireless Indices Report, at 85 – 90. 57 See Infra Footnote 22. 58 This ARPU is not equal to the ‘average bill’ for a household, or consumer, as it is not equal to an ‘account’ which may cover several different devices, such as multiple phones (under a family plan) or multiple devices (including phones and tablets, wireless broadband modems, or other adjunct devices covered by a customer’s service plan). It assigns overall service revenue across all revenue generating devices. See CTIA Year-End 2013 Wireless Indices Report, at 2. 59 See Appendix Table II.D.i 0 50 100 150 200 250 300 350 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 Year Chart II.D.1 Total Service Revenues , Subscribers , ARPU 1993 - 2013 Reported Subscribers (millions) Total Annual Service Revenue ($ billions) ARPU ($) 19 Federal Communications Commission DA 14-1862 providers, despite the overall stable numbers. From Table II.D.1 below we find that between the fourth quarter of 2011 and the fourth quarter of 2013, AT&T’s ARPU was fairly stable, Verizon Wireless experienced a slight increase in ARPU, and T-Mobile showed a steady decline in ARPU. Regional providers such as US Cellular experienced a slight decrease in ARPU during this time. Table II.D.1 ARPU Estimates of Publicly Traded Facilities-Based Mobile Wireless Providers 4th Quarter 2011 – 2nd Quarter 2014 Nationwide Providers 4Q11 4Q12 4Q13 2Q 14 AT&T $ 47.04 $ 46.94 $ 47.58 $ 43.91 Verizon Wireless $ 46.55 $ 47.57 $ 47.50 $ 46.96 Sprint $ 43.08 $ 43.37 $ 44.83 $ 43.55 T-Mobile $ 44.29 $ 40.24 $ 36.91 $ 36.17 Regional/Rural Providers 4Q11 4Q12 4Q13 US Cellular $ 49.74 $ 50.89 $ 50.21 $ 53.27 MetroPCS $ 40.55 $ 40.86 Leap Wireless $ 42.39 $ 40.69 $ 45.55 NTELOS $ 48.57 $ 52.78 $ 54.11 $ 52.21 Cincinnati Bell $ 43.26 $ 43.28 $ 41.35 $ 42.81 Source: UBS Investment Research. UBS Wireless 411 v. 51 Figure 36, UBS Wireless 411 Report Version 54. 39. As a consequence of the shift to shared data plans by the two largest wireless service providers, estimates of the unit price of wireless voice and data revenues are increasingly unreliable and difficult to come by, as discussed in earlier paragraphs. In addition, we note that the available estimates do not fully reflect the prices of all relevant mobile broadband services offered by U.S. wireless service providers, and therefore are subject to certain caveats depending on the methodology used in the particular analyst report. Acknowledging these limitations, we present some analyst estimates of ARPU and the unit price of mobile wireless broadband services. As seen in Table II.D.1, on average, the combined ARPU for voice, text and data has been fairly stable for most nationwide providers, with the exception of T-Mobile, which shows a steady decline in ARPU over the reported period. 3. Wireless Telephone Services CPI 40. The Consumer Price index (CPI) is a measure of the average change over time in the prices paid by urban consumers for a fixed market basket of consumer goods and services. The basket of goods includes over 200 categories, such as food and beverages, housing, apparel, transportation, medical care, recreation, education, and communications. The CPI allows consumers to compare the price of the basket of goods and services this month with the price of the same basket a month or a year ago. As documented in previous Reports, two different pricing indicators – the Wireless Telephone Services CPI and the per-minute price of voice service – show that mobile wireless prices have declined significantly since the launch of PCS service in the mid-1990s. However, given the shift in mobile voice service plans away from a defined number of monthly minutes60, there is no simple way to calculate a per-minute price for such service, so this discussion focuses on the CPI. 41. The wireless telephone services’ component of the CPI (Wireless Telephone Services CPI) is published by the U.S. Department of Labor’s Bureau of Labor Statistics (BLS) on a national basis.61 From 2011 60 See Section II.D.1 infra 61 Starting in December 1997, the basket included a category for cellular/wireless telephone services. All CPI figures discussed above were taken from BLS databases found at http://www.bls.gov. The index used in this analysis, the CPI for All Urban Consumers (CPI-U), represents about 87 percent of the total U.S. population. See Bureau of Labor Statistics, Consumer Price Index: Frequently Asked Questions, http://www.bls.gov/cpi/cpifaq.htm (visited June 16, 2014). The Cellular 20 Federal Communications Commission DA 14-1862 to 2012, the annual Wireless Telephone Services CPI decreased by 1.2 percent while the overall CPI increased by 2.1 percent and the Telephone Services CPI was approximately unchanged.62 From 2012 to 2013, the annual Wireless Telephone Services CPI decreased by 1.6 percent while the overall CPI increased by 1.5 percent and the Telephone Services CPI was unchanged. The Wireless Telephone Services CPI has steadily declined since 2010 following an unchanged Wireless Telephone Services CPI in 2009 and a series of much smaller declines in the period from 2002 to 2008. Since December 1997, the Wireless Telephone Services CPI has declined nearly 43 percent while the overall CPI has increased by 34 percent. 4. Profitability Metrics 42. One measure of competition is the relative profitability of competitors within the wireless market. It is also informative to compare the profitability of the wireless industry with other industries. In the absence of the data necessary to estimate economic profits, accounting profits can instead be estimated using various metrics available to wireless industry observers. One such metric, based on company data reported to the Securities and Exchange Commission, is EBITDA (Earnings before Interest, Taxes, Debt, and Amortization). EBITDA equals accounting profits before deducting interest expenses, corporate income taxes, depreciation, and amortization. In 2014, out of the nationwide facilities-based providers, EBITDA per subscriber ranged from a low of $6.13 (US Cellular) to a high of $24.19 (Verizon Wireless). These numbers are presented in Table II.D.2. Table II.D.2 Annual EBITDA per Subscriber ($/month), 2011 – 1st half 2014 Top 5 Mobile Wireless Service Providers 2011 2012 2013 1st Half 2014 Verizon Wireless 20.85 22.21 23.56 24.19 AT&T 18.49 18.64 19.55 19.67 Sprint 6.84 6.11 7.53 11.13 T-Mobile 13.17 12.09 10.08 8.64 US Cellular 11.88 11.51 7.34 6.13 Source: UBS Investment Research. UBS Wireless 411 Version 51, Fig 47; UBS Wireless 411 Version 54. Annual figures calculated by taking the average of each quarter for each year. 43. A second indicator of mobile wireless segment profitability is EBITDA margin63, which expresses EBITDA as a percentage of service revenue. Standardizing EBITDA by service revenues facilitates cross-provider comparisons. The EBITDA margin of a number of the publicly reported mobile providers for the past several years is shown in Chart II.D.1. In the fourth quarter of 2013, the EBITDA margin of the top four nationwide providers ranged from 47.0 percent (Verizon Wireless), to 14.4 percent (Sprint). The EBITDA margin of Verizon Wireless has remained above 40 percent since the fourth quarter of 2005.64 AT&T’s EBITDA margin has fluctuated since 2009, dropping below 30.0 percent in 2011, then rising above the 40 percent mark in late 2012 and in mid- 2103.65 The other providers’ EBITDA margins were all substantially lower in the second quarter of 2013. CPI includes charges from all telephone companies that supply “cellular telephone services,” which are defined as “domestic personal consumer phone services where the telephone instrument is portable and it sends/receives signals for calls by wireless transmission.” This measure does not include business calls, telephone equipment rentals, portable radios, and pagers. While the CPI-U is urban-oriented, it does include expenditure patterns of some of the rural population. Information submitted by companies for the CPI is provided on a voluntary basis. See Sixteenth Report, 28 FCC Rcd at 3836 at ¶ 263. 62 For details, see Appendix Table II.D.ii, 63 It is equal to earnings before interest, tax, depreciation and amortization (EBITDA) divided by total revenue. Because EBITDA excludes depreciation and amortization, EBITDA margin can provide a cleaner view of a company's core profitability. 64 UBS, US Wireless 411 Reports, 2002 – 2014. 65Id. 21 Federal Communications Commission DA 14-1862 Source: UBS Investment Research UBS Wireless 411 Version 51, Fig 46; UBS Wireless 411 Version 54. III. OVERALL MOBILE WIRELESS INDUSTRY METRICS 44. In this section, we discuss the current market trends in the mobile wireless marketplace and provide additional analysis highlighting specific changes that have occurred over the last year. Specifically, this section examines such indices as numbers of connections and distribution of subscribers by geography and by demographics. It analyzes the extent of voice and broadband coverage, including by number of available providers, and a comparison of rural to non-rural markets. We will also discuss developments in wireless devices, intermodal developments such as wireless-wireline substitution and wireless-only households, and consumers’ access to information about their available choices in the marketplace. A. Network Coverage 45. The analysis in this section is based on U.S. census blocks66 overlaid on provider coverage maps provided to the Commission through a contract with Mosaik Solutions, an independent consulting firm that tracks 66 A census block is the smallest geographic unit for which the Census Bureau tabulates decennial census data. There are 11,166,336 blocks designated in the 2010 Census, and they range in population from zero to several hundred. See U.S. Census Bureau, 2010 Census Summary File 1 – 2010 Census of Population and Housing, Technical Documentation, Mar. 2010, at 2-1, available at http://www.census.gov/prod/cen2010/doc/sf1.pdf#page=504. 1Q 11 2Q 11 3Q 11 4Q 11 1Q 12 2Q 12 3Q 12 4Q 12 1Q 13 2Q 13 3Q 13 4Q 13 1Q 14 2Q 14 Verizon Wireless 43.7 45.4 47.8 42.2 46.3 49 50 41.4 50.4 49.8 51.1 47 52.1 50.2 AT&T 39.7 41.7 44.3 29.2 42.3 45.8 41.6 29.1 43.2 42.4 42 37.4 45.9 42.6 Sprint 19.1 16.3 17.6 9.5 14.6 17.8 15.3 8.8 19.2 17.6 16.2 14.4 25.3 25.3 T-Mobile 25.7 27.6 31 30.7 28.7 30.5 28.8 24 28.9 24.7 26.2 24 20.3 26.4 Metro PCS 27.2 32.1 28.9 31.9 22.6 41.1 41.5 27.9 Leap (Cricket) 16.1 21.8 21.4 18 17 24.8 18 20.8 17.4 21.5 11.1 9.1 US Cellular 20.6 25.3 22.5 15.7 22.4 22.8 19.7 13.5 20.4 19.9 18.3 -7.8 9.27 11.1 -10 0 10 20 30 40 50 60 Pe rc en ta ge Chart II.D.1 Reported EBITDA Margins (in %) for Selected Publicly Traded Facilities-Based Wireless Providers 2011 - 1st Half 2014 22 Federal Communications Commission DA 14-1862 coverage footprints of mobile voice and mobile data networks. 67 If the center point, or centroid, of a census block is within the coverage boundary of a map provided by Mosaik, then we consider the census block to be “covered” by that provider and/or technology.68 We then aggregate the population and land area of the covered census blocks. These coverage estimates represent deployment of mobile networks and do not indicate the extent to which providers actually offer service to residents in the covered areas. While recognizing that this analysis likely overstates the coverage experienced by consumers because of limitations in Mosaik data, 69 we find that this analysis is useful because it provides a general baseline that can be compared over time across network technologies, and providers. 1. Overall Network Coverage 46. We first estimate the percentage of the U.S. population, land area, and road miles covered by a certain number of facilities-based mobile wireless service providers.70 We then present estimated mobile broadband coverage, using the same categories. For purposes of this Report, mobile wireless coverage represents either mobile voice or mobile broadband coverage, and “mobile broadband” includes coverage and services offered using the following 3G and 4G technologies: EVDO, EVDO Rev A, WCDMA/HSPA, HSPA+, LTE, and mobile WiMAX.71 Finally, we note the data underlying these estimates measure mobile network “coverage,” and not the number of network providers affirmatively offering service to any or all residents in those locations. a. Mobile Wireless Coverage 47. As of January 2014, 99.9 percent of the total U.S. population lived in census blocks that were covered by at least one facilities-based mobile wireless provider, as shown in Chart III.A.1. The percentage of the population living in a census block with mobile wireless coverage by at least one or two providers has not changed significantly since January 2012,72 while the percentage of the population living in a census block with coverage by at least three or four providers has fallen slightly during this time.73 However, the percentage of the population living in a census block covered by at least five providers fell sharply to 22.8 percent in January 2014, 67 Mosaik provides data to the FCC under contract on facilities-based providers in the form of coverage boundary maps based on the coverage boundaries provided to them by mobile wireless network operators. See Mosaik, About Us, http://www.mosaik.com/about-us/ (visited July. 7, 2014). 68 The Centroid Method overlays the geographic polygons showing wireless coverage onto a map of census blocks. The Centroid Method codes a census block as covered if the calculated center point (the “centroid”) of the census block is within the coverage polygon. If a centroid is covered, then all of the population and land area in the corresponding census block is coded as covered as well. We also note that in some cases the calculated center point may lay outside of the boundaries of a census block. In these cases, the centroid will be identified as the point inside the census block nearest to the calculated center point. 69 This analysis likely overstates the coverage actually experienced by consumers, because Mosaik reports advertised coverage as reported to it by many wireless service providers, each of which uses a different definition or determination of coverage. The data does not expressly account for factors such as signal strength, bit rate, or in-building coverage, and may convey a false sense of consistency across geographic areas and service providers but nonetheless are useful for benchmarking mobile network deployment across the United States, especially over time. National Broadband Plan, at 39 (Chapter 4). We also recognize that an analysis of coverage at the nationwide level provides only a general benchmark. A nationwide average will mask regional disparities in coverage and create an overall picture that does not capture variances across the country. 70 Also see Appendix Table III.A.i and III.A.ii 71 The Commission has used alternative definitions of mobile broadband in different contexts. For the 706 Report (Broadband Progress Report), it has used both a speed threshold based on the SBI data and a technology threshold based on the Mosaik data to define mobile broadband. See Eighth Broadband Progress Report, FCC Rcd 12-90A1 at 10366 ¶¶ 37 – 40. 72 Sixteenth Report, 28 FCC Rcd at ¶ 45 73 Sixteenth Report, 28 FCC Rcd at 3836, Table 4¶ 44 - 47 23 Federal Communications Commission DA 14-1862 from 79.6 percent in January 2012. We note that the number of providers in a census block does not necessarily reflect the number of network provider choices available to a particular individual or household residing in those areas.74 48. Chart III. A.1 also presents the approximate percentage of the U.S. land area and road miles covered by a certain number of mobile wireless providers. While more than 90 percent of the U.S. population lived in census blocks with coverage by at least four mobile voice providers in January 2014, these census blocks accounted for only approximately 29.9 percent of the total land area of the United States, and approximately 54.1 percent of U.S. road miles. Furthermore, while 0.1 percent of the U.S. population lived in census blocks that received no mobile wireless coverage, approximately 25 percent of the U.S. land area and 5 percent of U.S. road miles were not covered. Note: Based on January 2014 Mosaik Data and 2010 Census Data. The number of providers in a census block represent network coverage, which does not necessarily reflect the number of choices available to a particular individual or household, Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. 49. In this Report, we have included a provider if it has market share above a particular threshold, and have made estimates based on two alternative thresholds. Specifically, to estimate the number of providers serving a CMA, we include a provider if it has a greater than two percent market share (alternatively, a five percent market share which provides greater assurance of a meaningful choice for consumers) of mobile wireless connections based on NRUF data at the CMA level. Table III.A.1 presents the data for December 2013. Since the 74 The percentages of population located in census blocks with zero, one, two, or three or more mobile wireless or mobile broadband providers represent network coverage, which does not necessarily mean that they offered service to residents in the census block. In addition, we emphasize that a provider reporting mobile wireless or mobile broadband coverage in a particular census block may not provide coverage everywhere in the census block. For both these reasons, the number of providers in a census block does not necessarily reflect the number of choices available to a particular individual or household, and does not purport to measure competition. In addition, calculations based on Mosaik data on coverage, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless and mobile broadband coverage. 10.4% 29.9% 47.1% 64.9% 74.9% 18.4% 54.1% 72.9% 88.9% 95.5% 22.8% 91.4% 96.8% 99.4% 99.9% 0.0% 20.0% 40.0% 60.0% 80.0% 100.0% 5 or more 4 or more 3 or more 2 or more 1 or more Mobile Wireless Coverage (%) N um be r of S er vi ce P ro vi de rs w it h C ov er ag e (2 01 4) Chart III.A.1 Estimated Mobile Wireless Coverage by Census Block January 2014 % of U.S. Population % of U.S. Road Miles % of U.S. Square Miles 24 Federal Communications Commission DA 14-1862 Sixteenth Report,75 the percentage of CMAs with three or more providers have remained essentially unchanged from 29.3to 29.7 percent. 76 CMAs with 4 or more providers have increased from 34.4 percent to 43.7 percent, while there has been a decrease in the number of CMAs with at least five providers based on the five percent threshold primarily due to increased industry consolidation as discussed above. Table III.A.1 Estimated Mobile Wireless Providers Offering Service by CMA, Excluding Territories December 2013 Two Percent Market Share Threshold Five Percent Market Share Threshold Number of Providers Offering Service Anywhere in a CMA Number of CMAs Total CMAs (percent) Number of CMAs Total CMAs (percent) Total for U.S., excluding territories 716 100.0% 716 100.0% 1 provider 0 0.0% 2 0.3% 2 providers 62 8.7% 139 19.4% 3 providers 148 20.7% 213 29.7% 4 providers 358 50.0% 313 43.7% 5 or more providers 148 20.7% 49 6.8% Source: Based on December 2013 NRUF data. Just as is the case for census blocks, the number of providers in a CMA represents network coverage, which does not necessarily reflect the number of choices available to a particular individual or household. b. Mobile Broadband Coverage 50. Chart III.A.2 presents mobile wireless broadband coverage as of January 2014. 77 As discussed in an earlier section, for purposes of this Report, “mobile broadband” includes coverage and services offered using the following 3G and 4G technologies: EVDO, EVDO Rev A, WCDMA/HSPA, HSPA+, LTE, and mobile WiMAX.78 Mobile broadband coverage has generally increased since the Sixteenth Report.79 While this increase was small overall, there was significant expansion of specific broadband technologies, especially LTE, during this time.80 Despite the general increase in broadband coverage, the percentage of the U.S. population living in areas with five or more broadband providers fell to 11.8 percent in January 2014, compared to 52.9 percent in January 2012. 51. Mobile wireless broadband deployment focuses on high population centers. While more than 80 percent of the U.S. population lived in census blocks with coverage by at least four mobile service providers in January 2014, these census blocks only accounted for 35.3 percent of road miles, and 15.7 percent of the total land area of the United States. Furthermore, while 0.3 percent of the U.S. population lived in census blocks that received no mobile wireless broadband coverage, 7.3 percent of U.S. road miles and 20.8 percent of the U.S. land area were not covered. 75 See Sixteenth Report ¶ 50. 76 See Appendix Table III.A.iii for December 2011 data. Because NRUF includes data on the number of telephone numbers that have been assigned to end-user devices by mobile wireless providers, this analysis does not include providers whose data-only devices are not assigned a mobile telephone number. See also Section V.A, and Customers, infra. 77 Also see Appendix Table III.A.iv and III.A.v. 78 See Footnote 70 infra. 79 See Sixteenth Report, 28 FCC Red at 3836, ¶48 80 This is discussed in more detail in section VI.B of this Report, and in Table VI.B.1, below 25 Federal Communications Commission DA 14-1862 Note: Based on January 2014 Mosaik Data and 2010 Census Data. The number of providers in a census block represent network coverage, which does not necessarily reflect the number of choices available to a particular individual or household, Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. c. Urban/Rural Comparisons 52. Since the release of the Sixth Report,81 the Commission has also evaluated competition in rural areas. The Communications Act does not include a statutory definition of what constitutes a rural area.82 Since its 2004 Report and Order concerning deployment of wireless services in rural areas, the Commission has used a “baseline” definition of rural as a county with a population density of 100 persons or fewer per square mile.83 We 81 Sixth Report, 16 FCC Rcd at 13350. 82 The federal government has multiple ways of defining rural, reflecting the multiple purposes for which the definitions are used. See Eighth Report, 18 FCC Rcd at 14834; Facilitating the Provision of Spectrum-Based Service to Rural Areas and Promoting Opportunities for Rural Telephone Companies to Provide Spectrum-Based Services, See also Notice of Proposed Rulemaking, 18 FCC Rcd 20802, 20808-11 (2003). The Commission has used Rural Services Areas (RSAs) as a proxy for rural areas for certain purposes, such as the former cellular cross-interest rule and the former CMRS spectrum cap, stating that “other market designations used by the Commission for CMRS, such as [EAs], combine urbanized and rural areas, while MSAs and RSAs are defined expressly to distinguish between rural and urban areas.” See 1998 Biennial Regulatory Review, Spectrum Aggregation Limits for Wireless Telecommunications Carriers, Report and Order, 15 FCC Rcd 9219, 9256 ¶ 84, n.203 (1999). 83 Facilitating the Provision of Spectrum-Based Services to Rural Areas and Promoting Opportunities for Rural Telephone Companies To Provide Spectrum-Based Services, Report and Order, 19 FCC Rcd. 19078, 19087-88 (2004) (“We recognize, however, that the application of a single, comprehensive definition for ‘rural area’ may not be appropriate for all purposes. . . . Rather than establish the 100 persons per square mile or less designation as a uniform definition to be applied in all cases, we instead believe that it is more appropriate to treat this definition as a presumption that will apply for current or future Commission wireless radio service rules, policies and analyses for which the term ‘rural area’ has not been expressly defined. By doing so, we maintain continuity with respect to existing definitions of ‘rural’ that have been tailored to apply to specific policies, while also providing a practical guideline”). 2.5% 15.7% 33.8% 57.6% 70.2% 5.8% 35.3% 58.8% 82.9% 92.7% 11.8% 82.1% 93.4% 98.8% 99.7% 0.0% 20.0% 40.0% 60.0% 80.0% 100.0% 5 or more 4 or more 3 or more 2 or more 1 or more Mobile Wireless Broadband Coverage N um be r of S er vi ce P ro vi de rs w ith C ov er ag e (2 01 4) Chart III.A.2 Estimated Mobile Wireless Broadband Coverage by Census Block January 2014 % of U.S. Population % of U.S. Road Miles % of U.S. Square Miles 26 Federal Communications Commission DA 14-1862 use this same definition to analyze service availability in rural areas in this Report. By this definition, roughly 59 million people, or 19 percent of the U.S. population, live in rural counties, based on 2010 US Census data. These counties comprise 3.1 million square miles, or 86 percent of the geographic area of the United States.84 Approximately 81 percent of the U.S. population lives on 15 percent of the land, while 19 percent live on the remaining 85 percent of the land.85 (i) Mobile Wireless Network Coverage 53. As seen in Chart III.A.3, 100 percent of the non-rural population lived in census blocks that were covered by at least one provider in January 2014, compared to 99.3 percent of the rural population.86 The percentage of the population living in census blocks covered by at least two providers was also similar for rural and non-rural areas. As with mobile wireless coverage, the gap between rural and non-rural mobile broadband coverage jumps when we consider coverage by at least three or more providers. However, a higher percentage of the rural population lives in census blocks that were covered by at least five providers, with 25.9 percent of the rural population live in census blocks covered by at least five providers, compared to 20.8 percent of the non-rural population.87 While the percentage of rural and non-rural populations living in census blocks covered by at least one, at least two, at least three, and at least four providers has not changed significantly since January 2012, the percentage of the population living in census blocks with at least five providers has again decreased since January 2012.88 Source: Based on January 2014 Mosaik Data and 2010 Census Data. The number of providers in a census block represent network coverage, which does not necessarily reflect the number of choices available to a particular individual or household, Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have 84 Based on 2010 Census data. Includes the population of Puerto Rico. 85 Id 86 Also see Appendix Tables III.A.iv and III.A.v 87 This is largely a result of T-Mobile’s acquisition of MetroPCS and AT&T’s acquisition of Leap Wireless. Because the acquired providers had a larger presence in non-rural areas, the effect of the consolidation was more pronounced in those areas. 88 Sixteenth Report, 28 FCC Rcd at 3836, Table 55 and 57 25.9% 63.6% 84.2% 96.4% 99.3% 20.8% 96.2% 99.2% 99.9% 100.0% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 5 or more 4 or more 3 or more 2 or more 1 or more Percentahe of US POPs N um be r of S er vi ce P ro vi de rs w ith C ov er ag e (J an ua ry 2 01 4) Chart III.A.3 Percentage of Population Living in a Census Block Covered by Mobile Wireless Voice Providers in Rural vs. Non-Rural Areas , January 2014 Non-Rural (% of U.S. POPs) Rural (% of U.S. POPs) 27 Federal Communications Commission DA 14-1862 certain limitations that likely overstate the extent of mobile wireless coverage 54. Chart III.A.4 presents mobile wireless coverage of rural and non-rural road miles in January 2014.89 The changes in road mile coverage since January 2012 are similar to the trends in population coverage over the same time period. Source: Based on January 2014 Mosaik Data and 2010 Census Data. The number of providers in a census block represent network coverage, which does not necessarily reflect the number of choices available to a particular individual or household, Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. (ii) Mobile Broadband Network Coverage 55. As seen in Chart III.A.5, approximately 100 percent of the non-rural population lived in census blocks that were covered by at least one provider in January 2014, compared to 98.5 percent of the rural population. 90 The percentage of the population living in census blocks covered by at least two providers was also similar for rural and non-rural areas. The gap between rural and non-rural coverage jumps when we consider coverage by at least three or more providers, narrowing again when we consider the percentage of the population living in census blocks covered by at least five providers. The percentage of rural residents living in census blocks with at least one available mobile broadband provider did not change significantly between January 2012 and January 2014.91 During the same time period, the percentage of rural residents living in census blocks with at least two, at least three, or at least four available mobile broadband providers increased by 8.4 percent, 13.6 percent, and 9.9 percent, respectively. Meanwhile, the percentage of rural residents living in census blocks with at least five available mobile broadband providers fell only slightly, from 10.3 percent to 8.6 percent The percentage of non-rural residents living in census blocks with at least one, two, three, or four available mobile broadband providers increased only slightly during this time, while the percentage of non-rural residents living in census blocks with at least five broadband providers dropped significantly, from 62.8 percent to 12.5 percent, largely reflecting the acquisitions of MetroPCS and Leap/Cricket by T-Mobile and AT&T, respectively. 89 Also see Appendix Tables III.A.vi and III.A.vii 90 Also see Appendix Tables III.A.viii and III.A.ix 91 Sixteenth Report, 28 FCC Rcd at 3836, Tables 55 and 57, ¶385-387 13.0% 35.1% 58.8% 81.2% 92.5% 22.4% 82.2% 93.3% 97.7% 99.3% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 5 or more 4 or more 3 or more 2 or more 1 or more Percentage of US Road Miles N um be r of S er vi ce P ro vi de rs w ith C ov er ag e (J an ua ry 2 01 4) Chart III.A.4 Percentage of Road Miles Covered by Mobile Wireless Voice Providers in Rural vs. Non-Rural Areas, January 2014 Non-Rural (% of U.S. Road Miles) Rural (% of U.S. Road Miles) 28 Federal Communications Commission DA 14-1862 Source: Based on January 2014 Mosaik Data and 2010 Census Data. The number of providers in a census block represent network coverage, which does not necessarily reflect the number of choices available to a particular individual or household, Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage 56. Chart III.A.6 presents mobile broadband coverage of rural and non-rural road miles in January 2014.92 The percentage of rural road miles with mobile broadband coverage increased across the board between January 2012 and January 2014. The percentage of non-rural roads covered by at least one, at least two, at least three, and at least four providers increased, however the percentage of non-rural roads covered by at least five providers fell dramatically from 40.5 percent to 10.8 percent between January 2012 and January 2014. Source: Based on January 2014 Mosaik Data and 2010 Census Data. The number of providers in a census block represent network coverage, which does not necessarily reflect the number of choices available to a particular individual or household, Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have 92 Also see Appendix Tables III.A.viii and II.A.ix 8.6% 39.6% 72.0% 94.4% 98.5% 12.5% 92.0% 98.4% 99.8% 100.0% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 5 or more 4 or more 3 or more 2 or more 1 or more Percentahe of US POPs N um be r of S er vi ce P ro vi de rs w ith C ov er ag e (J an ua ry 2 01 4) Chart III.A.5 Percentage of Population Living in a Census Block Covered by Mobile Broadband Providers in Rural vs. Non-Rural Areas , January 2014 Non-Rural (% of U.S. POPs) Rural (% of U.S. POPs) 3.3% 17.7% 44.0% 76.2% 89.7% 10.8% 71.5% 89.4% 96.9% 98.8% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 5 or more 4 or more 3 or more 2 or more 1 or more Percentage of US Road Miles N um be r of S er vi ce P ro vi de rs w it h C ov er ag e (J an ua ry 2 01 4) Chart III.A.6 Percentage of Road Miles Covered by Mobile Broadband Providers in Rural vs. Non-Rural Areas , January 2014 Non-Rural (% of U.S. Road Miles) Rural (% of U.S. Road Miles) 29 Federal Communications Commission DA 14-1862 certain limitations that likely overstate the extent of mobile wireless coverage. 2. Network Coverage and Roaming 57. Service providers often use roaming services to enhance their coverage. They offer their customers coverage outside of their network coverage areas through roaming arrangements with other providers. Roaming arrangements between mobile wireless service providers allow customers of one mobile wireless provider to automatically receive service from other providers’ networks when they are in areas that are covered by their roaming partners’ networks but not their own network. Smaller providers that rely on roaming arrangements to offer nationwide coverage to their customers often include the price of nationwide roaming services in the plans’ monthly fees instead of billing for roaming on a usage basis. In contrast to the purchase of capacity wholesale from other service providers to provide resale or MVNO services, a provider uses roaming services to market extended coverage to consumers residing within the provider’s network coverage area, not to acquire customers where a provider does not have network coverage. A detailed discussion is provided in Section VI.B, which discusses non-price rivalry between providers. 3. Network Coverage by Technology 58. The Commission has adopted flexible licensing policies, and does not mandate any particular technology or network standard for commercial mobile wireless licensees. Mobile wireless service providers choose their own network technologies and services and abide by certain technical parameters designed to avoid radiofrequency interference with adjacent licensees. As a result of this approach, over the past 15 years U.S. service providers have deployed a variety of digital network technologies with divergent technology migration paths. Previously, two main technology migration paths have been the CDMA and GSM paths.93 There has not been any significant change in CDMA or GSM/TDM coverage since the Sixteenth Report, and each technology covers over 99 percent of the population. Sprint’s iDEN network was shut down on July 2013, and the percentage of the population that is covered by this technology accordingly fell from 90.0 percent in January 2012, to 4.9 percent in January 2014 with service being provided by small providers in a few markets.94 The evolution of mobile network technologies is now converging on LTE, as all of the major service providers are deploying or planning to deploy LTE technology.95 59. During the time period covered by this Report, the four nationwide facilities-based mobile wireless service providers, as well as other mobile providers continued to upgrade and expand their networks with advanced 3G and 4G technologies that allow for faster mobile broadband connection speeds.96 LTE, in particular, has been growing in importance over the past few years, as it can provide faster speeds and improved user experience. Each provider is extending its LTE footprint in order to better compete in the mobile wireless marketplace. According to Verizon Wireless, about 69 percent of total data traffic was over the 4G/LTE network in.97 As of January 2014, 98.5 percent of the population lived in census blocks that were covered by an LTE 93 See Appendix Chart III.A.i. Of the top four nationwide mobile wireless providers, AT&T and T-Mobile have deployed technologies on the GSM migration path, while Verizon Wireless and Sprint have deployed technologies on the CDMA migration path. Sprint has shut down its iDEN network. 94 Appendix Table III.A.viii presents mobile wireless network coverage by technology type in terms of population, land area, and road miles, as of January 2014 95 See section VII.B.3 for a more detailed discussion of service provider network deployments. 96 For purposes of this Report, the term “broadband” – when referring to mobile broadband networks, coverage, providers, or services – includes the 3G and 4G network technologies: HSPA, EV-DO, LTE, and mobile WiMAX. The Commission may include other combinations of mobile network technologies when referring to “mobile broadband” in other contexts. See, e.g., Eighth Broadband Progress Report at Table 15. 97 VZ – Q4 2013 Verizon Earnings Conference Call, January 21, 2014 30 Federal Communications Commission DA 14-1862 network, compared to 67.5 percent of the population in January 2012. 98 WCDMA/HSPA/HSPA+ coverage also increased during this time, increasing from 93.1 percent of the population in January 2012 to 97.7 percent of the population in January 2014 as shown in Table III.A.2. Table III.A.2 Estimated Mobile Wireless Data/Broadband Network Coverage by Census Block, Jan. 201499 Technology POPs in Covered Blocks (Thousands) % of Total POPs Square Miles Contained in Those Blocks (Thousands) % of Total Square Miles Road Miles Contained in Those Blocks (Thousands) % of Total U.S. Road Miles 2.5G CDMA 1xRTT 310,365 99.3% 2,532 66.6% 6,117 89.7% GPRS/EDGE 310,396 99.3% 2,507 65.9% 6,082 89.2% Total 2.5G Mobile Data Network Coverage 311,962 99.8% 2,788 73.3% 6,460 94.7% 3G/ 4G WCDMA/HSPA/ HSPA+ 305,138 97.7% 2,121 55.8% 5,421 79.5% EV-DO/EV-DO Rev. A 310,024 99.2% 2,434 64.0% 6,001 88.0% Mobile WiMAX 105,486 33.8% 44 1.2% 419 6.1% LTE 307,736 98.5% 2,067 54.4% 5,475 80.3% Total Mobile Broadband Coverage (3G/4G) 311,492 99.7% 2,669 70.2% 6,322 92.7% Note: Based on January 2014 Mosaik Data and 2010 Census Data. Calculations based on Mosaik data on coverage, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile broadband coverage. The number of providers in a census block reflect network coverage, which does not necessarily reflect the number of choices available to a particular individual or household 4. Network Coverage by Income Levels 60. We also analyze how the number of facilities-based mobile wireless providers that have coverage in a census tract varies based on median household income levels.100 The analysis is based on mobile wireless and 98 The analysis of mobile wireless network coverage in this section is based on U.S. census blocks overlaid on provider coverage maps provided to the Commission through a contract with Mosaik Solutions, described above. Also see CR 16 table 31 99 Includes Federal lands. Commission estimates based on census block analysis of Mosaik CoverageRight coverage maps, January 2014. Population data are from the 2010 Census, and the square miles include the United States and Puerto Rico. 100 The percentages of population located in census tracts where zero, one, two, or three or more mobile wireless or mobile broadband providers represent network coverage, which does not necessarily mean that they offered service to residents in the census block. In addition, we emphasize that a provider reporting mobile wireless or broadband coverage in a particular census tract may not provide coverage everywhere in the census tract. For both these reasons, the number of providers in a census tract, or by income level does not necessarily reflect the number of choices available to a particular individual or household at a certain income level, and does not purport to measure competition. In addition, calculations based on Mosaik data on coverage, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless and mobile broadband coverage. 31 Federal Communications Commission DA 14-1862 mobile broadband coverage data reported by Mosaik101 and the median household income levels in each of the country’s 74,000 census tracts based on United States Census Bureau’s American Community Survey (ACS).102 Chart III.A.7 below shows that, as of January 2014, the average number of mobile wireless providers in census tracts with median household income less than $25,000 was 4.65, compared to 4.10 in census tracts with median household income of more than $150,000. The average number of mobile broadband providers in census tracts with median household income less than $25,000 was 4.41, compared to 3.95 in census tracts with median household income of more than $150,000. Chart III.A.8 compares the number of mobile broadband providers by income level for August 2010, January 2012, and January 2014.103 Source: Data on median household income by census tract are based on United States Census Bureau’s American Community Survey 2009-2013 (ACS). Data on number of providers are from Mosaik, January 2014. The number of mobile wireless or mobile broadband providers in a census tract represents network coverage, which does not necessarily mean that they offered service to any or all the residents in the census tract. In addition, we emphasize that a provider reporting mobile wireless or broadband coverage in a particular census tract may not provide coverage everywhere in the tract. 101 Data on numbers of mobile wireless providers and mobile broadband providers are based on Mosaik database, January 2014. We note that the calculations based on Mosaik data on coverage, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless and mobile broadband coverage. 102 Data on median household income are based on United States Census Bureau’s American Community Survey 2009-2013 (ACS). The analysis is done on a census tract, rather than census block, basis because the smallest geographic area for which median household income data is available is census tracts. These data do not allow for an analysis of adoption rates for mobile wireless or mobile broadband services. 103 See Sixteenth Mobile Wireless Competition Report, 28 FCC Rcd ¶ 297, for the August 2010 and January 2012 data. 4.65 4.29 4.42 4.47 4.45 4.44 4.10 4.41 3.93 4.15 4.35 4.38 4.39 3.95 0.00 1.00 2.00 3.00 4.00 5.00 N um be r of W ir el es s P ro vi de rs a nd W ir el es s B ro ad ba nd P ro vi de rs Median Household Income by Census Tract Chart III.A.7 Average Number of Mobile Wireless Providers and Mobile Broadband Providers in Census Tracts by Median Household Income January 2014 Average Number of Mobile Wireless Providers Average Number of Mobile Broadband Providers 32 Federal Communications Commission DA 14-1862 Source: Current data on median household income by census tract are based on United States Census Bureau’s American Community Survey 2009-2013 (ACS). Current data on number of providers are from Mosaik, January 2014. August 2010 and January 2012 data are from the Sixteenth Competition Report (Chart 41). The number of mobile wireless or mobile broadband providers in a census tract represents network coverage, which does not necessarily mean that they offered service to any or all the residents in the census tract. In addition, we emphasize that a provider reporting mobile wireless or broadband coverage in a particular census tract may not provide coverage everywhere in the tract. 5. Commission Actions Related to Coverage, Technology and Roaming 61. When competing mobile wireless service providers deploy compatible network technologies, greater economies of scale in the production of both end-user devices and network infrastructure equipment can result, lowering the unit cost of handsets, chipsets, and other network equipment. This, in turn, may promote more rapid adoption of mobile wireless services, a greater variety of handsets, and more price competition. In October, 2013, the Commission adopted a Report and Order and Order of Proposed Modification to effectuate a voluntary industry agreement and thereby provide for interoperable LTE service in the Lower 700 MHz band. Since that time, the Commission has adopted specific interoperability requirements for the AWS-3 band, as well as for the 600 MHz Band.104 104 See Amendment of the Commission’s Rules with Regard to Commercial Operations in the 1695-1710 MHz, 1755-1780 MHz, and 2155-2180 MHz Bands, GN Docket No. 13-185, Report and Order, 29 FCC Rcd. 4610, ¶¶ 225-231 Rel. March 31, 2014, (AWS-3 Report and Order) and Expanding the Economic and Innovation Opportunities of Spectrum Through Incentive Auctions, GN Docket No. 12-268, Report and Order, 29 FCC Rcd. 6567, ¶¶ 731-737 Rel. June 2, 2014 (Incentive Auctions Report and Order). 3.32 3.26 3.71 3.74 3.7 3.75 3.7 4.85 4.56 5.04 5.32 5.36 5.42 5.47 4.41 3.93 4.15 4.35 4.38 4.39 3.95 0 1 2 3 4 5 6 N um be r of W ir el es s B ro ad ba nd P ro vi de rs Median Household Income by Census Tract Chart III.A.8 Average Number of Mobile Broadband Providers in Census Tracts by Median Household Income August 2010, January 2012, January 2014 Aug-10 Jan-12 Jan-14 33 Federal Communications Commission DA 14-1862 62. In order to encourage mobile network deployment into unserved or underserved areas, the Commission adopted rules creating the Mobility Fund in November 2011.105 The Mobility Fund uses Universal Service Fund reserves to support the deployment of current- or future-generation mobile network technologies that provide mobile voice and Internet services.106 For Mobility Fund Phase I, the Commission will provide up to $300 million in one-time support payments, plus up to $50 million dedicated to Tribal lands, that were awarded through reverse auctions,.107 The Commission is currently exploring whether to retarget Mobility Fund Phase II ongoing support to ensure the continued deployment and preservation of 4G LTE mobile broadband service and preservation of mobile voice and broadband service in areas that otherwise would not have such service through marketplace forces.108 63. In recent years, the Commission has taken actions to facilitate roaming arrangements.109 In 2007, for instance, it clarified that automatic voice roaming is a common carrier obligation for CMRS providers.110 In April 2010, the Commission adopted the Roaming Order on Reconsideration, which eliminates the home roaming exclusion and establishes the same general obligation to provide automatic voice roaming, regardless of whether the provider requesting roaming holds spectrum in an area.111 In April 2011, the Commission issued the Data Roaming Order.112 The Data Roaming Order requires facilities-based providers of commercial mobile data services, whether or not such providers also offer CMRS, to offer data roaming arrangements to other mobile data service providers on commercially reasonable terms and conditions, subject to certain limitations.113 64. Several commenters in the current record contend that it is still difficult to negotiate roaming agreements with larger, nationwide providers.114 A recent survey by NTCA of its membership, which consists exclusively of small, rural providers, asked participants to categorize their experience in negotiating data roaming and in-market roaming agreements with other providers. Of the respondents, 69 percent categorized it as moderately to extremely difficult, 27 percent as moderately to relatively easy, and four percent as extremely easy. In addition, 52 percent of those respondents who have a reciprocal roaming agreement with 105 Connect America Fund, A National Broadband Plan for Our Future, Establishing Just and Reasonable Rates for Local Exchange Carriers, High-Cost Universal Service Support, Developing an Unified Intercarrier Compensation Regime, Federal- State Joint Board on Universal Service, Lifeline and Link-Up, Universal Service Reform – Mobility Fund, Report and Order and Further Notice of Proposed Rulemaking, 26 FCC Rcd 17663 (2011). See Section IX, Urban-Rural Comparisons, infra. 106 Id 107 Auction 901 Closing Public Notice. Mobility Fund Phase I disbursements were authorized beginning April 2013 and are anticipated to continue through 2016. Mobility Fund Phase I Support Authorized for Seven Winning Bidders; Defaults on Two Auction 901 Winning Bids Determined, AU Docket No. 12-25, Public Notice, 28 FCC Rcd 5599. Tribal Mobility Fund Phase I Auction Closes Winning Bidders Announced for Auction 902, Public Notice, released February 28, 2014, XX FCC Rcd Commission. 108 See In the matter of Connect America Fund Universal Service Reform – Mobility Fund, ETC Annual Reports and Certifications, Establishing Just and Reasonable Rates for Local Exchange Carriers, Developing an Unified Intercarrier Compensation Regime, WC Docket No. 10-90, Further Notice of Proposed Rulemaking, (rel. June 10, 2014). 109 Sixteenth Report, 28 FCC Rcd at 3837 ¶ 209. 110See Roaming Obligations of Commercial Mobile Radio Service Providers, Report and Order and Further Notice of Proposed Rulemaking, 22 FCC Rcd 15817, 15828 ¶ 27 (2007) (2007 Roaming Order and FNPRM) (“[W]e recognize that automatic roaming benefits mobile telephony subscribers by promoting seamless CMRS service around the country, and reducing inconsistent coverage and service qualities.”) 111Roaming Order on Reconsideration, 25 FCC Rcd at 4182 ¶ 2. 112Data Roaming Order, 26 FCC Rcd 5411. 113Id. at 5418-5428 ¶ 13-31. 114 See. e.g. CCA Comments at 17. 34 Federal Communications Commission DA 14-1862 another provider indicated that they pay about as much as they themselves are paid, while 33 percent pay more and 14 percent pay less.115 65. On May 27, 2014, T-Mobile USA, Inc. (“T-Mobile”) filed a petition for an expedited declaratory ruling that would provide guidance on the criteria used for determining whether the terms of a data roaming agreement meet the “commercially reasonable” standard set forth in the Commission’s data roaming rule.116 The data roaming rule requires facilities-based providers of commercial mobile data services to offer roaming arrangements to other such providers on “commercially reasonable terms and conditions.”117 T-Mobile contends that providers need this guidance to evaluate the commercial reasonableness of terms offered in individual negotiations and to reach agreements. The Commission released a Public Notice on June 10, 2014 seeking comment on the petition.118 B. Connections and Subscribers 1. Connections and Subscribers by Geography 66. To better understand the number of connections across geographic areas, for this Report, we have estimated penetration rates,119 using NRUF subscriber data,120 at the level of the 172 Economic Areas (EAs)121 of the United States, each of which is an aggregation of a differing number of counties. We use EAs as the geographic unit for measuring the level of concentration in the mobile wireless services marketplace in order to maintain continuity with past Reports and to ensure that we do not compromise the confidential information contained in the NRUF data.122 Regional penetration rates for the 172 EAs range from 85 percent in La Crosse, WI-MN to 188 percent in Grand Island, NE.123 The nationwide penetration rate based on NRUF data now exceeds 100 percent, meaning that the number of connected devices exceeds the population, and the penetration 115 NTCA 2012 Wireless Survey Report, September 2012, at 3. See http://www.ntca.org/images/stories/Documents/Advocacy/SurveyReports/2012ntcawirelesssurveyreport.pdf 116 Petition for Expedited Declaratory Ruling of T-Mobile USA, Inc., WT Docket No. 05-265, filed May 27, 2014. 117 47 C.F.R. § 20.12(e). 118 Wireless Telecommunications Bureau Seeks Comment on Petition for Expedited Declaratory Ruling by T-Mobile USA, Inc. Regarding Data Roaming Obligations, WT Docket No. 05-265, Public Notice, DA 14-798 (WTB rel. June 10, 2014). 119 The penetration rate is defined as the number of mobile wireless connections per 100 people. 120 NRUF subscriber data indicate the number of assigned phone numbers that a wireless provider has in a particular wireline rate center (there are approximately 18,000 rate centers in the country). Rate centers are geographic areas used by local exchange carriers for a variety of reasons, including the determination of toll rates. See HARRY NEWTON, NEWTON’S TELECOM DICTIONARY: 19TH EXPANDED & UPDATED EDITION 660 (July 2003). All mobile wireless providers must report to the Commission the quantity of their phone numbers that have been assigned to end users, thereby permitting the Commission to calculate the total number of mobile wireless subscribers. For purposes of geographical analysis, the rate center data can be associated with a geographic point, and all of those points that fall within a county boundary can be aggregated together and associated with much larger geographic areas based on counties. We note that the aggregation to larger geographic areas reduces the level of inaccuracy inherent in combining non-coterminous areas such as rate center areas and counties. 121 EAs are geographic areas defined by the U.S. Department of Commerce that define geographic markets using data on commuting patterns. . We recognize that EAs may be broader or narrower than other geographic markets employed in the Commission’s analyses. For example, the Commission typically has used smaller geographic areas, such as CMAs, in its analysis of mobile wireless transactions. See, e.g., Sprint Nextel-Clearwire Order, 23 FCC Rcd at 17591 ¶¶ 51-52; Verizon Wireless-Alltel Order, 23 FCC Rcd at 17472-73 ¶ 52. 122 See Section II.D infra. 123 See Appendix Table III.B.i 35 Federal Communications Commission DA 14-1862 rate in 82 of the 172 EAs was at least 100 percent at the end of 2013.124 2. Connections and Subscribers by Demographics 67. Several socio-economic and demographic factors such as household income and age are correlated with overall mobile wireless subscription rates as well as smartphone subscription rates. Based on August 2014 survey data from ComScore Mobilens,125 Chart III.B.1 shows that mobile wireless subscribers overall, and smartphone subscribers in particular, are in higher income brackets. For example, 24.7 percent of the population live in households with an annual income of less than $25,000, but only 16.5 percent of mobile wireless users and 13.1 percent of smartphone users are in this bracket. Conversely, 22.0 percent of the population live in households with an annual income over $100,000, but 28.2 percent of mobile wireless subscribers and 32 percent of smartphone subscribers are in this income bracket. The chart also shows that income may also be correlated with the choice of a prepaid plan or a postpaid plan: more postpaid users are in a higher income bracket, while the converse is true for prepaid subscribers. Source: ComScore, MobiLens Audience Profile August 2014 3-month survey data averages and U.S. Census Bureau 2012 Population Reports. 68. The ComScore data also allows the presentation of the composition of mobile users by age. The age distribution of mobile wireless subscribers and of smartphone subscribers is shown in Chart III.B.2. While the general adoption of mobile wireless devices is fairly evenly distributed among various age groups, smartphone adoption is more concentrated in younger age groups. For example, adults ages 18-44 comprise 47 percent of all mobile wireless subscribers, but make up over 55 percent of smartphone users, while adults over the age of 55 124 According to the Bureau of the Census, the combined population of the 50 states, the District of Columbia, and Puerto Rico, as of July 1, 2011, was estimated to be 311.6 million. See U.S. Census Bureau, http://www.census.gov/popest/data/national/totals/2011/index.html (visited June 17, 2011). As noted in the Fifteenth Report, if NRUF is used to calculate a mobile wireless penetration rate (of a population), that penetration rate is overstated in terms of the number of individuals who have more than one mobile wireless device. 125 Survey data based on ComScore MobiLens, March 2014. ComScore MobilLens U.S. data are derived from a monthly survey of over 13,000 respondents ages 13 and older who are recruited to represent U.S. Census demographics. The total universe size is estimated from data provided by CTIA and comScore’s monthly subscriber studies. Race data are found at the US Census Bureau State and County Quick Facts 2013 website, at http://quickfacts.census.gov/qfd/states/00000.html. Income data are found in the “Income, Poverty, and Health Insurance Coverage in the United States: 2012 Current Population Reports. United States Census Bureau”. Issued September 2013, Table A-1: Households by Total Money Income, Race, and Hispanic Origin of Householder, at http://www.census.gov/prod/2013pubs/p60-245.pdf 16.5% 13.1% 10.6% 27.8% 21.4% 19.6% 19.0% 26.5% 19.8% 20.0% 21.0% 18.4% 14.2% 15.3% 16.0% 10.8% 28.2% 32.0% 33.4% 16.5% U.S. Mobile Subscribers U.S. Smartphone Subscribers Postpaid Subscribers Prepaid Subscribers Pe rc em t o f Su bs cr ib er s Chart III.B.1 Total Mobile Subscribers, Smartphone Subscribers, and Prepaid/Postpaid Subscribers by Income Bracket, August 2014 $100k+ $75k to <$100k $50k to <$75k $25k to <$50k <$25k *CTIA - 335.6 million. *NRUF - 336.0 million 36 Federal Communications Commission DA 14-1862 represent over 29 percent of all mobile wireless subscribers, but only 21 percent of smartphone subscribers.126 Source: ComScore MobiLens, 3 Month Average, October 2014 C. Consumers and Mobile Wireless 69. In today’s connected world, consumers are faced with a wide variety of choices in mobile service plans, devices and applications. But fundamental to these options is the choice of a mobile service provider. Consumers choose a service provider or switch between providers for varying reasons, including price, availability of family plans, network quality, free/unlimited in-network calling, billing/payment options/credit, reputation/recommendation, previous experience with the provider, customer service, mobile data services, specific phone offerings, and bundling mobile phone services with other services or other unspecified reasons. In the past, contract length, handset exclusivity, lack of interoperability were some factors that were highlighted as barriers to switching. Recently, the advent of no-contract plans, such as those discussed in Section V, newer premium models such as the new iPhone versions being available to more providers, and the FCC 700 MHz interoperability Order, may have eased some of the switching barriers, and somewhat reduced switching cost. However, even now, switching is not free of costs. When mobile wireless customers wish to switch service providers, they may incur some switching costs including: search costs; early termination fees (ETFs); handset purchase; and implicit costs such as brand loyalty. 1. Usage 70. According to CTIA, reported annual MOUs increased 13.8 percent, reaching over 2.6 trillion. Average billable minutes of use (MOUs), a measure of monthly mobile voice usage per connection, also increased significantly in 2013.127 This follows a decline in average MOUs, which leveled out in 2012. As seen in Chart III.C.1, between 2012 and 2013, average monthly MOUs, excluding most data-only devices, increased by 10.5 percent, compared with a previous year-over-year decline of 0.4 percent from 2011 to 2012 and 6.3 percent decline from 2010 to 2011. According to CTIA research staff, this may be due in part to improved reporting, provider participation and possible volume increases in usage, the 2013 MOUs show a significant increase in MOUs reported to CTIA.128 127 CTIA Wireless Industry Indices Year-End 2013 at page 129, Table 43. 128 CTIA Wireless Industry Indices Year-End 2013 at page 135. Telephone Conversation between FCC staff and CTIA Research October 2014. 12.9% 15.5% 17.4% 21.0% 16.7% 18.8% 17.7% 17.1% 14.9% 12.1% 14.1% 8.4% 0.0% 20.0% 40.0% 60.0% 80.0% 100.0% U.S. Mobile Subscibers U.S. Smartphone Subscribers P er ce nt o f S ub sc ri be rs Chart III.B.2 Total Mobile Subscribers Compared to Smartphone Subscribers, by Age October 2014 Age 65+ Age 55-64 Age 45-54 Age 44-35 Age 25-34 Age 18-24 37 Federal Communications Commission DA 14-1862 Source: CTIA Wireless Industry Indices Year-End 2013. Latest available data. 71. However, voice usage does not tell the entire mobile use story. Cisco projects that mobile data will grow at an annual rate of 50 percent from 2013 to 2018129 while Ericsson—a network infrastructure provider—projects mobile data growth of 38 percent per year between 2013 and 2019.130 This trend is due to multiple factors including increased adoption of smartphones and tablets, growth in streaming video, and the development of faster networks. 72. CTIA reported that SMS and text messaging traffic amounted to over 153.3 billion for the December 2013 period. According to the CTIA survey, average monthly data usage per subscriber in 2013 averaged 1.2 GB per month increasing 50 percent over 2012. Chart III.C.2 provides average data usage per subscriber for 2010 to 2013 comparing the amount of data usage between data-capable devices and smartphones. Source: CTIA Wireless Industry Indices Year-End 2013, Chart 32 Indices. Latest available data. 73. Other research organizations such as NPD indicates that according to 2013 data, average mobile data usage rangles between 550 MB and 1.4 GB.131 According to GSMA, LTE users use twice as much data as non-LTE users , which translates to about 1.5 GB of data per month on average.132 Total wireless data traffic 129 Cisco Visual Networking Index: Forecast and Methodology, 2013-2018, (June 10, 2014). 130Cisco Visual Networking Index; Ericsson, Ericsson Mobility Report; On the Pulse Of the Networked Society (June 2014). 131 http://www.fiercewireless.com/special-reports/average-android-ios-smartphone-data-use-across-tier-1-wireless-carriers- thr-1 132 GSMA Report, The Mobile Economy, 2014. http://www.gsmamobileeconomy.com/GSMA_ME_Report_2014_R2_WEB.pdf 758 717 672 675 746 726 683 650 633 690 550 600 650 700 750 800 2009 2010 2011 2012 2013 A ve ra ge M O U s pe r Su bs cr ib er p er M on th Year Chart III.C.1 Average MOU per Subscriber per Month, 2009-2013 Excludes most data-only devices Includes non-voice capable units 122 257 407 849 269 510 768 1152 0 500 1000 1500 2010 2011 2012 2013 A ve ra ge M on th ly M B p er A ve ra ge D at a C ap ab le U ni t Chart III.C.2 Mobile Data Usage per Subscriber, 2010 - 2013 Data Capable Unit Smartphone 38 Federal Communications Commission DA 14-1862 reported by the providers to CTIA amounted to 3.23 trillion MB for 2013 up 120 percent from 1.47 trillion MB in 2012.133 For the third quarter of 2014, Mobidia reports that LTE continues to drive data usage with the average 3G smartphone subscriber using less than half the data of and LTE subscriber, who average monthly data usage is around 2 GB.134 74. According to the Pew Research Internet Project, 81 percent of cellphone users use their cellphone to send or receive text messages; 60 percent access the Internet; 52 percent send or receive email; 50 percent download apps; 49 percent get directions, recommendations, or other location-based information; 48 percent listen to music; 21 percent participate in a video call or video chat; and 8 percent check-in or share location.135 Not only has the variety of uses changed, but socially acceptable times to use mobile devices have also changed. For example, 70 percent of mobile device users reported having, within the previous 24 hours, used a mobile device while eating.136 2. Handsets 75. Smartphone Penetration. Smartphone use has continued to increase over the last two years. The Pew Research Internet Project estimates that as of January 2014, 90 percent of all American adults had a cell phone, and 58 percent had a smartphone.137 Chart III.C.3 presents a more detailed analysis. According to ComScore’s dataset, 72 percent of all mobile subscribers had a smartphone in September 2014,, compared to 51 percent in September 2012. 138 These numbers increase when we consider only subscribers who purchased a phone recently. For instance, 85 percent of subscribers purchasing a new phone in September 2014 were smartphone users, up from 67 percent in September 2012. 133 CTIA Indices at page 11 134 Mobidia LTE Data Usage, November 2014, http://www.mobidia.com/blog 135 http://pewinternet.org/Reports/2013/Cell-Activities.aspx 136 Citrix Mobile Device Survey, January 2014. 137 Pew Research Internet Project, Cell Phone and Smartphone Ownership Demographics, Jan. 2014, available at http://www.pewinternet.org/data-trend/mobile/cell-phone-and-smartphone-ownership-demographics/ 138 ComScore, MobiLens Trend, 3 month averages from August 2012 to November 2013. 39 Federal Communications Commission DA 14-1862 Note: ComScore, MobiLens Audience Profile, 3 month averages from August 2012 to September 2014. 76. Share of Smartphones by Operating System. The operating system of a smartphone is a major determining factor of the smartphone’s ability to support mobile applications and Internet-based services. As seen in Chart III.C.4139, Apple’s iOS and Google’s Android continued to dominate the market for mobile operating systems.140 In September 2014 Android’s share of the market was 52 percent, and it retained over half of the smartphone operating system market. In second place, Apple’s iOS held 42 percent of the market in September 2014, up from 34.3 percent in September 2012. RIM (Research in Motion)/Blackberry (2 percent), Microsoft (4 percent), and other firms (1 percent) comprised the remainder of the market. Note: Based on ComScore MobiLens 3-month survey data averages 139 ComScore, MobiLens Trend. comScore MobilLens U.S. data are derived from a monthly survey of over 13,000 respondents ages 13 and older who are recruited to represent U.S. Census demographics. The total universe size is estimated from data provided by CTIA and comScore’s monthly subscriber studies. 140 Also see Appendix Table III.C.i 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% A ug -2 01 2 Se p- 20 12 O ct -2 01 2 N ov -2 01 2 D ec -2 01 2 Ja n- 20 13 Fe b- 20 13 M ar -2 01 3 A pr -2 01 3 M ay -2 01 3 Ju n- 20 13 Ju l- 20 13 A ug -2 01 3 Se p- 20 13 O ct -2 01 3 N ov -2 01 3 D ec -2 01 3 Ja n- 20 14 Fe b- 20 14 M ar -2 01 4 A pr -2 01 4 M ay -2 01 4 Ju n- 20 14 Ju l- 20 14 A ug -2 01 4 Se p- 20 14 P en et ra tio n R at e (% ) Month-Year Chart III.C.3 Smartphone Penetration Rates in the US (3 Month Average) August 2012 - September 2014 Of all Mobile Phones Of Recent Purchases (Last Month) 5% 20% 45% 52% 52% 52% 25% 24% 27% 34% 41% 42% 42% 38% 19% 8% 4% 2% 18% 11% 6% 4% 3% 4% 10% 8% 3% 1% 1% 1% 0% 10% 20% 30% 40% 50% 60% December 2009 August 2010 September 2011 September 2012 September 2013 September 2014 M ar ke t S ha re ( % ) Chart III.C.4 Market Share by Smartphone Model (2009-2014) Google/Android Apple RIM/BlackBerry Microsoft All Others 40 Federal Communications Commission DA 14-1862 3. Mobile Applications 77. The increasing use of smartphones has spawned a mobile applications ecosystem. Major categories include: web searching, news and information, e-mail and messaging, games, social networking, location-based services, photo sharing, music and video streaming, VoIP, and mobile commerce (including mobile payments, mobile banking, and mobile shopping). Thousands of niche applications have been designed for specific uses, hobbies, interests, and industries by various third-party application developers. The number of mobile applications launched and the number of applications downloaded by consumers have grown significantly over the past three years. There are two main application stores—the Apple App Store and Google Play. As of year end 2013, based on revenues, the Apple App Store had a 62 percent market share, while Google Play had a 38 percent market share.141 78. Additionally, mobile commerce is playing a growing role in the U.S. economy.142 As of November 2013, approximately 36 percent of U.S. bank account holders have used mobile banking services more than once in the past 30 days.143 Mobile banking allows consumers to check account balances, pay bills, and transfer funds on a variety of mobile devices.144 Many banks offer consumers text banking, access to accounts via the mobile web, mobile banking applications, and mobile deposits for use on several platforms and devices.145 In addition, many mobile wireless handsets and devices can be used to make on-the-spot payments at physical retail locations with a technology commonly known as “Tap and Pay.” Mobile payments technologies include SMS, operator billing, the mobile Internet, mobile wallets, and Near Field Communications (NFC).146 As of, February 2014, approximately 87 percent of smartphone and tablet owners say they use these devices for shopping activities.147 79. Most mobile applications are available for download through mobile web browsers or through mobile application stores, such as the Apple’s App Store or Google Play. Once an application is installed on a mobile device, the application may or may not require a mobile broadband connection to function. In addition, many applications for smartphones and other devices such as tablets are pre-installed on mobile operating systems. As see in Chart III.C.5, as of July 2014, Android users were able to choose between 1.3 million apps. 141 Top Global Apps – January 2014, http://www.distimo.com/blog/2014_02_top-global-apps-january-2014/, (visited 5/22/2014). 142 Matthew B. Gross, Jeanne M. Hogarth, and Maximilian D Schmeiser, Consumers and Mobile Financial Services, Federal Reserve Board, Division of Consumer and Community Affairs, March 2012. Penny Crosman, Contactless Mobile Pay Transactions Seen Nearing 10B by 2016, Mar. 6, 2012. 143 Nielsenwire, The Nielsen Company, Multiplying Mobile: How Multicultural Consumers are Leading Smartphone Adoption, March 4, 2014. Neilsen Newswire, The Nielsen Company, The Evolution of Modern Banking, March 19, 2014. 144 International Business Times, Mobile Banking on the Rise, Aug. 15, 2012, available at http://www.ibtimes.com/articles/373841/20120815/mobile-banking-united-states-bank-account-holders.htm (visited April 16, 2014). 145 See generally Bank of America, Mobile Banking. http://www.bankofamerica.com/onlinebanking/index.cfm?template=mobile_banking (visited April 16, 2014); Chase, Chase Mobile Banking, https://www.chase.com/index.jsp?pg_name=ccpmapp/shared/assets/page/Chase_Mobile_Banking (visited April 16, 2014); Citibank, Citi Mobile Banking, https://online.citibank.com/US/JRS/pands/detail.do?ID=CitiMobile (visited April 16, 2014). 146 International Business Times, Mobile Banking on the Rise, Aug. 15, 2012, available at http://www.ibtimes.com/articles/373841/20120815/mobile-banking-united-states-bank-account-holders.htm (visited April 16, 2014); Ovum, Mapping Mobile Payments, April 2012, at 2. 147 Nielsenwire, The Nielsen Company, Shopping Lists: How Mobile Helps Consumers Tick All the Boxes, February 20, 2014. 41 Federal Communications Commission DA 14-1862 Apple's App Store remained the second-largest app store with 1.2 million available.148 Note: Data from Statistica.com, Latest available data. 80. Mobile applications are available in a broad range of categories and include web searching, social media, and gaming. As shown in Chart III.C.6, the applications that were accessed by the highest percentage of smartphone users in April 2014 were email, weather, and social networking apps.149 However, other apps continue to grow in popularity. Analysts predict that industry-focused mobile applications marketplaces will develop, focusing for example on professional healthcare or education, among other industries.150 81. Mobile applications generate revenue through contracts for application developers, e-commerce sales, in-application advertising, and application store sales.151 Estimates from Vision Mobile and Developer Economy indicate that the total worth of the global mobile applications marketplace ranged from $60 billion to $70 billion in 2013.152 148 See http://www.statista.com/statistics/276623/number-of-apps-available-in-leading-app-stores/ 149 See ComScore, MobiLens. 3 month average survey data, April 2014. 150 See Vision Mobile, Business and Productivity Apps, March 2014. 151 See Vision Mobile, Business and Productivity Apps, March 2014. 152 See Developer Economics, State of the Developer Nation, February 2014. See also, Vision Mobile, Business and Productivity Apps, March 2014. 0 500000 1000000 1500000 Google Play Apple App Srore Windows Phone Store Amazon App Store Blackberry World N um be r of M ob il e A pp s Chart II.C.5 Number of Mobile Apps Available in Leading App Stores July 2014 42 Federal Communications Commission DA 14-1862 Note: Based on ComScore MobiLens 3-month survey data averages, August 2014 4. Consumer Access to Information 82. Through the Consumer Code for Wireless Service, CTIA and the service providers that are its signatories voluntarily commit to providing consumers with information to assist them in the selection of a mobile wireless service provider.153 Signatories to CTIA’s Consumer Code commit to disclose rates, additional taxes, fees, surcharges, and terms of service; provide coverage maps; and make customer service readily accessible. In July 2010, CTIA updated the Consumer Code to require providers to ensure disclosure of data allowances offered in a service plan, whether there are any prohibitions on data service usage, and whether there are network management practices that will have a material impact on the customer’s wireless data experience.154 The Consumer Code also states that prepaid service providers must disclose the period of time during which any prepaid balance is available for use.155 Some wireless service providers have implemented formal procedures to permit consumers to use their service on a trial basis for periods ranging from 14 to 30 days, consistent with one of the elements of CTIA’s Consumer Code.156 83. Bill Shock. In October 2011, CTIA revised its Consumer Code to require that its participating providers provide four types of alerts (data, voice, text, international roaming) by April 17, 2013, and at least two out of the four types of alerts by October 17, 2012.157 The member providers agreeing to this plan account for service to 97 percent of U.S. wireless customers and all customers are included unless they opt out. In order to 153 See CTIA, Consumer Code for Wireless Service, available at http://files.ctia.org/pdf/ConsumerCode.pdf (visited Oct. 16, 2012). (Consumer Code for Wireless Service). 154 See CTIA Comments at 46; CTIA, CTIA-The Wireless Association® Announces Updates to Its ‘Consumer Code for Wireless Service,’ Press Release, July 28, 2010, available at http://www.ctia.org/media/press/body.cfm/prid/1992 (visited Oct. 16, 2012). 155 Id. 156 See CTIA Comments at 44-45; See also Consumer Code for Wireless Service. The ability of consumers to terminate a wireless service contract within 14 days is also one of a number of provisions of the Assurance of Voluntary Compliance agreed to by AT&T (then Cingular), Sprint Nextel, and Verizon Wireless with the attorneys general of 32 states on June 25, 2004. 157 See http://www.fcc.gov/blog/new-fcc-website-help-consumers-beat-%E2%80%98bill-shock%E2%80%99 (visited Oct. 16, 2012). See also CTIA Consumer Code, http://www.ctia.org/content/index.cfm/AID/10352 (visited Nov. 1, 2012). 0.0% 20.0% 40.0% 60.0% 80.0% 100.0% Pe rc en t o f U se rs Top Categories of Mobile Apps and Websites Chart III.C.6 Top 15 Types of Mobiles Apps and Websites Accessed by Smartphone Users in August, 2014 43 Federal Communications Commission DA 14-1862 further facilitate the adoption of such alerts, the Commission has established a web site where consumers can determine which providers are implementing the voluntary commitments.158 84. Open Internet Rules. The rules on Internet openness adopted by the Commission in December 2010 require both fixed and mobile broadband Internet providers to “publicly disclose accurate information regarding the network management practices, performance, and commercial terms of its broadband Internet access services sufficient for consumers to make informed choices regarding use of such services.”159 In providing guidance regarding effective disclosure models in that Order, the Commission indicated that among the types of information that might be included in an effective disclosure are pricing terms such as monthly prices, usage-based fees, and fees for early termination or additional network services.160 The Commission also adopted anti-blocking requirements for fixed and mobile providers and an anti-discrimination rule for fixed providers.161 85. Verizon challenged the open Internet rules in the D.C. Circuit and the court ruled on Verizon’s challenge in January, 2014.162 The court rejected Verizon’s challenge to the transparency rule, but struck down the anti-blocking and anti-discrimination rules. The court remanded the case to the Commission for further proceedings. In May, 2014, the Commission issued a NPRM responding to the court’s remand, and proposed to adopt new rules consistent with the court’s opinion.163 Among its proposals, the Commission tentatively concluded that it should enhance the transparency rule to improve its effectiveness and require broadband providers to more specifically tailor disclosures to the needs of affected parties.164 5. Intermodal Developments 86. We here provide the latest information from the CDC National Health Interview Survey on wireless-only households. Preliminary results from CDC’s July - December 2013 National Health Interview Survey indicate that the number of American homes with only wireless telephones continues to grow. As shown in Chart III.C.7 the percentage of U.S. adults and children living in households with landlines, with or without wireless, has fallen steadily over the past few years.165 The percentage of wireless-only households has continued to increase for both groups, and the percentage of households without phones has not changed significantly. 166 However, a significant percentage of homes with both landline and wireless phone access received all or almost all calls on wireless telephones despite also having a landline telephone. 167 SNL Kagan, estimates that phone cord cutting will continue to grow over the next decade, driving wireless-only households to 60.9 percent of all 158 See http://www.fcc.gov/encyclopedia/bill-shock-wireless-usage-alerts-consumers (visited Oct. 16, 2012). 159 Open Internet Order at ¶ 54. 160 Open Internet Order at ¶ 56. 161 47 C.F.R. § 8.5. 162 Verizon v. FCC, 740 F.3d 623 (D.C. Cir. 2014). 163 Protecting and Promoting the Open Internet, GN Docket No. 14-28, Notice of Proposed Rulemaking, 29 FCC Rcd 5561 (2014). 164 Id. at 5586, ¶¶ 67-68. 165 Also see Appendix Tables III.C.ii and III.C.iii 166 Stephen J. Blumberg and Julian V. Luke, Wireless Substitution: Early Release of Estimates from the National Health Interview Survey, January-June 2013, National Center for Health Statistics, Centers for Disease Control, December 2014, available at http://www.cdc.gov/nchs/data/nhis/earlyrelease/wireless201312.pdf 167 Stephen J. Blumberg and Julian V. Luke, Wireless Substitution: Early Release of Estimates from the National Health Interview Survey, January-June 2013, National Center for Health Statistics, Centers for Disease Control, December 2013, available at http://www.cdc.gov/nchs/data/nhis/earlyrelease/wireless201312.pdf 44 Federal Communications Commission DA 14-1862 households by the end of 2024.168 Source: Data from CDC/NCHS National Health Interview Surveys, Jan-June 2012 and Jan-June 2013. Latest available data. Adults are aged 18 and over, children are under age 18. IV. INPUT MARKETS 87. Mobile wireless service providers employ a combination of inputs to provide mobile wireless services to their customers. These inputs include electromagnetic spectrum to transmit signals between base stations and end users’ devices, as well as non-spectrum inputs such as cellular base stations and towers to carry transmissions. Backhaul, which routes voice and data traffic from base stations for onward transmission and may use spectrum or wireline resources, is an additional input required for the provision of mobile service. A. Spectrum 88. This section highlights the role that spectrum plays as an input in the provision of mobile wireless services, summarizes the Commission’s policies to facilitate the use of commercial wireless spectrum, and provides summary information on service providers’ current spectrum holdings. 1. Importance of Spectrum for the Provision of Mobile Wireless Services 89. As the Commission has recently found, spectrum is a critical input in the provision of mobile wireless services, including mobile broadband, as it affects if and when existing service providers and potential entrants will be able to expand capacity or deploy networks.169 Incumbent licensees may need additional spectrum to increase their coverage or capacity as they grow their subscriber bases and meet increasing demand, while new entrants need access to spectrum to enter the market and compete with incumbent licensees.170 90. Spectrum bands vary in their propagation characteristics, which has implications for spectrum use and deployment. Service providers deploy their spectrum bands differently depending on the nature of the 168 SNL Kagan (cite) 169 See Policies Regarding Mobile Spectrum Holdings; Expanding the Economic and Innovation Opportunities of Spectrum Through Incentive Auctions, WT Docket No. 12-269; Docket No. 12-268, Report and Order, FCC 14-63, at ¶ 2 (re. Jun. 2, 2014) (“Mobile Spectrum Holdings Report and Order”); Sixteenth Report, 28 FCC Rcd at 3836 ¶ 85. 170 Sixteenth Mobile Wireless Competition Report, 28 FCC Rcd 3769 ¶ 86. 63.7% 63.5% 62.5% 62.2% 59.4% 58.8% 57.3% 56.1% 54.4% 52.8% 51.5% 18.4% 21.1% 22.9% 24.9% 27.8% 30.2% 32.3% 34.0% 36.5% 38.0% 39.1% 15.1% 13.4% 12.6% 10.9% 10.7% 9.0% 8.3% 7.8% 7.0% 6.9% 7.0% 1.7% 1.5% 1.7% 1.7% 1.8% 1.8% 1.9% 1.9% 1.9% 2.2% 2.2% 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% Jul–Dec 2008 Jan–Jun 2009 Jul–Dec 2009 Jan–Jun 2010 Jul–Dec 2010 Jan–Jun 2011 Jul–Dec 2011 Jan–Jun 2012 Jul–Dec 2012 Jan–Jun 2013 Jul-Dec 2013 P er ce nt Chart III.C.7 Percentage of U.S. Adults Living in Households with/without Wireless and Landlines (2008 - 2013) Landline with Wireless Wireless-only Landline without Wireless Phoneless Landline with Wireless Wireless only Landline without Wireless Phoneless 45 Federal Communications Commission DA 14-1862 service, geography, density, or other factors in their network build-out.171 Spectrum below 1 GHz (“low-band spectrum”) has distinct propagation advantages for network deployment over long distances, while also reaching deep into buildings and urban canyons. Spectrum above 1 GHz (“high-band spectrum”) is more plentiful and possesses certain technical advantages allowing for the transmission of large amounts of information. In this sense, spectrum below 1 GHz may be thought of as “coverage” spectrum, and spectrum above 1 GHz may be thought of as “capacity” spectrum.172 There is significantly less low-band spectrum than high-band spectrum that is suitable and available for the provision of mobile telephony/broadband services.173 91. Rising consumer demand for mobile broadband is increasing service providers’ need for spectrum at an unprecedented rate.174 As service providers deploy next-generation mobile networks, the engineering properties and deployment capabilities of the mix of particular spectrum bands in their spectrum holdings have become increasingly important, particularly as multi-band phones allow users to take advantage of the different properties of different spectrum bands.175 Service providers need access to spectrum that can provide both coverage and in-building penetration, as well as spectrum that can provide the increased throughput for mobile broadband applications.176 A service provider holding a mix of low- and high-band spectrum licenses has greater flexibility and is better able to optimize its network costs for a given quality level.177 92. As the Commission has found, robust competition depends critically upon the availability of spectrum as a necessary input in the provision of mobile wireless services.178 For robust competition to exist and persist, multiple competing service providers must have access to a sufficient mix of low- and high-band spectrum to be able to enter a marketplace or expand output rapidly in response to any price increase or reduction in quality, or other change that would harm consumer welfare.179 In particular, without access to low-band spectrum, service providers would have to rely on alternative, less cost-effective methods to increase rural and in-building coverage to serve additional customers, such as adding towers, splitting cells, or acquiring roaming rights on other networks.180 In that regard, spectrum acquisition can be valuable in furthering a service provider’s competitive 171 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6149-6154 ¶ 31-40; Sixteenth Report, 28 FCC Rcd at 3789-3793 ¶ 119-127. 172 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6135 ¶ 3; Sixteenth Mobile Wireless Competition Report, 28 FCC Rcd at 3789, 3792 ¶ 121, 126. 173 Sixteenth Mobile Wireless Competition Report, 28 FCC Rcd at 3789, 3792 ¶ 121, 126. 174 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6134 ¶ 2. 175 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6144 ¶ 18; Sixteenth Mobile Wireless Competition Report, 28 FCC Rcd at XXXX ¶ 127. 176 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6157 ¶ 47;Sixteenth Mobile Wireless Competition Report, 28 FCC Rcd at XXXX ¶ 127. 177 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6163-6164 ¶ 59; Sixteenth Mobile Wireless Competition Report, 28 FCC Rcd at 3789, 3792-93, 3796 ¶¶ 119, 127, 135. 178 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6134 ¶ 1; Verizon Wireless-SpectrumCo Order, 27 FCC Rcd at 10716 ¶ 47; AT&T-Qualcomm Order, 26 FCC Rcd at 17601-02 ¶ 30; Verizon Wireless-ALLTEL Order, 23 FCC Rcd at 17481-82 ¶ 75; Cingular-AT&T Wireless Order, 19 FCC Rcd at 21569 ¶ 109. 179 Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6143, 6163-6164 ¶¶ 17, 59. See also AT&T WCS Order, 27 FCC Rcd at 16467 ¶ 20; Verizon Wireless-SpectrumCo Order, 27 FCC Rcd at 10716 ¶ 48; AT&T-Qualcomm Order, 26 FCC Rcd at 17601-02 ¶ 30. 180 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6164 ¶ 60. While other cost-related factors exist, ensuring that multiple providers are able to access a sufficient amount of low-band spectrum is a threshold requirement for extending and improving service in both rural and urban areas. See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6135 ¶ 3. 46 Federal Communications Commission DA 14-1862 position as well as reducing opportunities available to its rivals.181 2. Facilitating Access to Spectrum 93. Recognizing the importance of spectrum in the provision of mobile wireless services, Congress, through the Communications Act, requires the Commission to implement spectrum policies that promote competition, innovation, and the efficient use of spectrum to best serve the public interest, convenience and necessity.182 Consistent with the Commission’s statutory mandate, the Commission has established policies to make spectrum available to existing mobile service providers and potential new entrants through initial licensing, primarily by competitive bidding, and through secondary market transactions.183 The Commission’s spectrum policies have been guided by the goal of promoting and preserving competition, which in turn enables consumers to make choices among numerous service providers and leads to lower prices, improved quality, and increased innovation.184 The Commission generally has provided licensees with significant flexibility to decide which services to offer and what technologies to deploy on spectrum used for the provision of mobile wireless services, which has permitted an evolution to next-generation wireless technologies and services using the licensees’ existing spectrum. a. Auctions 94. Since 1994, the Commission has conducted various auctions of spectrum licenses.185 These auctions are open to any eligible entity that submits an application and upfront payment, and is found to be a qualified bidder by the Commission.186 The Sixteenth Report discusses auctions for the various frequency bands which are potentially suitable for the provision of mobile telephony/broadband service.187 Additionally, the Commission’s auction website provides detailed information regarding ongoing, completed, and planned auctions.188 95. To meet rising the consumer demand discussed above, the Commission is making substantially more spectrum available for the provision of mobile wireless services. In early 2014, the Commission auctioned the 10 megahertz of H Block in the 1.9 GHz Band, in which Dish Network won all 176 licenses.189 The auction 181 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6164, 6165 ¶¶ 60, 62. 182 See 47 U.S.C. § 309(j)(3)(B). 183 See, Promoting Efficient Use of Spectrum Through Elimination of Barriers to the Development of Secondary Markets, WT Docket No. 00-230, Second Report and Order, Order on Reconsideration, and Second Further Notice of Proposed Rulemaking, 19 FCC Rcd 17503, ; 700 MHz Second R&O, 22 FCC Rcd at 15374-80, ¶¶ 231-248 184 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6144 ¶ 17. Our public interest evaluation necessarily encompasses the “broad aims of the Communications Act,” which include, among other things, a deeply rooted preference for preserving and enhancing competition in relevant markets, accelerating private sector deployment of advanced services, and generally managing the spectrum in the public interest. See, e.g., AT&T WCS Order, 27 FCC Rcd at 16464 ¶ 11; AT&T- Centennial Order, 24 FCC Rcd at 13928 ¶ 28; Verizon Wireless-ALLTEL Order, 23 FCC Rcd at 17461 ¶ 27; Sprint- Clearwire Order, 23 FCC Rcd at 17580 ¶ 20. 185 See Auction 1 in FCC Auctions Home, Auctions, http://wireless.fcc.gov/auctions/default.htm?job=auctions_home (visited Jul. 03, 2014). 186 See Federal Communications Commission, About Auctions, http://wireless.fcc.gov/auctions/default.htm?job=about_auctions (visited Mar. 27, 2014). 187 See Sixteenth Report, 28 FCC Rcd at 3773-3778 ¶ 92-100. 188 See Federal Communications Commission, Auctions Home, http://wireless.fcc.gov/auctions/default.htm?job=auctions_home (visited Mar. 27, 2014). 189 See Auction of H Block Licenses in the 1915-1920 MHz and 1995-2000 MHz Bands Closes; Winning Bidder Announced for Auction 96, 29 FCC Rcd at 2044 (Re. Feb. 28, 2014); Service Rules for Advanced Wireless Services in the 2000-2020 MHz and 2180-2200 MHz Bands, WT Docket No. 12-70, Fixed and Mobile Services in the Mobile Satellite Service Bands at 47 Federal Communications Commission DA 14-1862 for Advanced Wireless Services-3 (“AWS-3”), which commenced on November 13, 2014, and is ongoing as of the release of this report, will award 65 megahertz of high-band spectrum to the winning bidders.190 In addition, the planned 600 MHz Incentive Auction will auction significant amounts of low-band spectrum in a broadcast television spectrum incentive auction. Pursuant to the Spectrum Act, the Commission will allow broadcasters to voluntarily participate in a “reverse auction” of the UHF spectrum, which would then be made available in a “forward auction” for licenses with flexible use service rules. It is expected that the 600 MHz Band will be used to provide robust mobile broadband service, given that its technical characteristics and rules are similar to those for the 700 MHz Band, which is the home today of much of the current LTE mobile broadband service. 191 96. In the Mobile Spectrum Holdings Report and Order, the Commission concluded that, in lieu of a post-auction application of the spectrum screen to the initial licensing of spectrum to winning bidders, the Commission would determine whether a band-specific mobile spectrum holding limit is necessary, and if so, would establish an ex ante application of that limit to the competitive bidding for that band. 192 With respect to the Incentive Auction, the Commission established a market-based spectrum reserve of up to 30 megahertz in each license area that is designed to ensure against excessive concentration in holdings of low-band spectrum while including safeguards to ensure that all bidders bear a fair share of the cost of the Incentive Auction. 193 The Commission declined to adopt band-specific mobile spectrum holding limits for AWS-3, emphasizing the availability of a substantial amount of comparable high-band spectrum to competitors and the significant existing holdings of multiple providers of comparable spectrum. 194 b. Secondary Markets 97. Subject to the Commission’s approval, licensees may assign and exchange licenses, in whole or in part (through partitioning and/or disaggregation), on the secondary market.195 In reviewing proposed acquisitions of spectrum through secondary market transactions, the Commission uses an initial screen to help identify for case-by-case review local markets where changes in spectrum holdings resulting from the transaction may be of particular concern.196 As set out in various transactions orders, however, the Commission has not 1525-1559 MHz and 1626.5-1660.5 MHz, 1610-1626.5 MHz and 2483.5-2500 MHz, and 2000-2020 MHz and 2180-2200 MHz, ET Docket No. 10-142, Service Rules for Advanced Wireless Services in the 1915-1920 MHz, 1995-2000 MHz, 2020- 2025 MHz and 2175-2180 MHz Bands, WT Docket No. 04-356, Report and Order and Order of Proposed Modification, FCC 12-151 (rel. Dec. 17, 2012) (AWS-4 Report and Order). 190 See Amendment of the Commission’s Rules with Regard to Commercial Operations in the 1695-1710 MHz, 1755-1780 MHz, and 2155-2180 MHz Bands, GN Docket No. 13-185, Report and Order, GN 13-185 (re. Mar. 31, 2014) (AWS-3 Report and Order). See also Auction of Advanced Wireless Services (AWS-3) Licenses Scheduled for November 13, 2014, AU- Docket No. 14-78, Public Notice, (rel. July 23, 2014). 191 See generally Expanding the Economic and Innovation Opportunities of Spectrum Through Incentive Auctions, GN Docket No. 12-268, FCC 14-50, Report and Order (May 15, 2014). 192 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6192 ¶ 139 (replacing policies previously articulated in 2008 in Union Telephone Company and Cellco Partnership d/b/a Verizon Wireless Applications for 700 MHz Band Licenses, Auction No. 73, Order, 23 FCC Rcd 16787, 16791, 16796 ¶¶ 9, 18 (2008)). See Section IV.A.2.b infra 193 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6193-6219 ¶¶ 146-217. 194 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6220-6221 ¶¶ 222-224. 195 As part of its secondary market policies, the Commission also permits mobile wireless licensees to lease all or a portion of their spectrum usage rights for any length of time within the license term, and over any geographic area encompassed by the license. For a more comprehensive overview of the Commission’s secondary market policies, see Sixteenth Report, 28 FCC Rcd at 3782-3783 ¶¶ 108-110. 196 See, e.g., Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6221-6222 ¶ 225; Alaska Wireless Order, 28 FCC Rcd at 10449-50 ¶ 38; Verizon Wireless-SpectrumCo Order, 27 FCC Rcd at 10719 ¶ 59; AT&T-Qualcomm Order, 26 FCC Rcd at 17602 ¶ 31; AT&T-Centennial Order, 24 FCC Rcd at 13938 ¶ 50. For transactions that result in the acquisition of wireless business units and customers or change the number of firms in any market, the Commission also applies an initial 48 Federal Communications Commission DA 14-1862 limited its consideration of potential competitive harms solely to markets identified by its initial screen, if it encounters other factors that may bear on the public interest inquiry.197 98. The Commission includes in its initial screen spectrum that it finds is suitable and available for the provision of mobile telephony/broadband services. Suitability is based upon whether the spectrum band at issue is capable of supporting mobile service given its physical properties and the state of equipment technology, whether the spectrum is licensed with a mobile allocation and corresponding service rules, and whether the spectrum is committed to another use that effectively precludes its uses for the relevant mobile services.198 With respect to availability, the Commission considers particular spectrum to be a relevant input if it is fairly certain that it will meet the criteria for suitable spectrum in the near term.199 99. In the past decade, in the context of its review of secondary market transactions, the Commission periodically determined that additional spectrum was suitable and available, and therefore subject to inclusion in the spectrum screen used in its competitive review—including 700 MHz,200 AWS-1,201 BRS,202 and WCS.203 Recently, in the Mobile Spectrum Holdings Report and Order, the Commission updated the spectrum screen by adding 151 megahertz of spectrum in total from the AWS-4 (2.0/2.2 GHz), H Block (1.9 GHz), BRS, and EBS bands.204 It also designated for future inclusion in the spectrum screen, the amount of 600 MHz Band spectrum that would be made available through the upcoming Incentive Auction, and the 65 megahertz of AWS-3 spectrum as it becomes available on a market by market basis.205 Furthermore, the Commission subtracted 12.5 megahertz of SMR, and 10 megahertz that was the Upper 700 MHz D Block.206 Spectrum currently included in the screen is as follows: screen based on the size of the post-transaction HHI and the change in the HHI. See, e.g., Cingular-AT&T Wireless Order, 19 FCC Rcd at 21564-65 ¶ 96. 197 See, e.g., Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6221-6222 ¶ 225; SoftBank-Sprint Order, 28 FCC Rcd at 9656 ¶ 35; AT&T WCS Order, 27 FCC Rcd at 16467 ¶ 21; Verizon Wireless-SpectrumCo Order, 27 FCC Rcd at 10716 ¶ 48; AT&T-Qualcomm Order, 26 FCC Rcd at 17610-11 ¶¶ 49-50. 198 See, e.g., Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6169 ¶ 71; AT&T-WCS Order, 27 FCC Rcd at 16469-16970 ¶ 29 and n. 81. 199 Id. 200 See AT&T-Dobson Order, 22 FCC Rcd 20295, 20307-08 ¶ 17. 201 See Sprint Nextel-Clearwire Order, 23 FCC Rcd 17570, 17599 ¶ 72. 202 Most BRS spectrum is considered available in those markets where the transition of BRS spectrum to the new band plan has been completed. Sprint Nextel-Clearwire Order, 23 FCC Rcd at 17598-99 ¶ 70; Amendment of Part 27 of the Commission’s Rules to Govern the Operation of Wireless Communications Services in the 2.3 GHz Band, Report and Order, 25 FCC Rcd 11710, 11711 ¶ 1 (2010). 203 See AT&T-WCS Order, 27 FCC Rcd at 16470-16471 ¶ 31; WCS Order on Reconsideration, 27 FCC Rcd at 13688 ¶ 88. 204 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6172-6187 ¶¶ 82-125. 205 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6171-6172, 6176-6179 ¶¶ 76-81, 94-102. 206 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6187-6190 ¶¶ 126-134. 49 Federal Communications Commission DA 14-1862 Table IV.A.1 Spectrum Included in the Spectrum Screen Spectrum Band Megahertz 700 MHz 70 Cellular 50 SMR (Specialized Mobile Radio Service) 14 Broadband PCS 130 AWS-1a 90 H-Block 10 AWS-4 40 WCS 20 BRSb (Broadband Radio Service) 67.5 EBS (Educational Broadband Service) 89 Total Amount of Spectrum 580.5 a AWS-1 is not attributable in markets where Federal Government users have not been relocated. b BRS is not attributable in markets where previous BRS licensees have not been transitioned. 100. For those markets identified by the spectrum screen, or where the Commission encounters other factors that may bear on the public interest inquiry, 207 the Commission conducts further competitive review to determine whether the transaction would result in an increased incentive or ability for the assignee or transferee to behave in an anticompetitive manner. The case-by-case analysis considers variables that are important in predicting the incentives and ability of service providers to successfully reduce competition on price or non-price terms, and transaction-specific public interest benefits that may mitigate or outweigh any public interest harms that might arise from the transaction.208 In addition, the Mobile Spectrum Holdings Report and Order found that considering additional below-1-GHz spectrum concentration as an enhanced factor in the Commission’s review of secondary market transactions will help ensure that further concentration of such spectrum will not have adverse competitive effects either in particular local markets or on a broader regional or national level. The Commission can condition approval of a transaction on the divestiture of licenses or certain other commitments in markets where necessary to find an application serves the public interest.209 101. Since the Sixteenth Report, a number of transactions involving the transfer of spectrum licenses, as well as, in certain cases, network infrastructure and other assets, have been filed with the Commission. Major transactions have included the transfer of control and assignments of various spectrum licenses of Atlantic Tele- Network, Inc.210 as well as Leap Wireless211 to AT&T; the transfer of control and assignments of various 207 For example, the Commission also considered whether harms in numerous local markets may result in nationwide harms and has considered potential harms from concentration in a particular band with an important ecosystem and from concentration of below-1-GHz spectrum. See, e.g., Verizon Wireless-SpectrumCo Order, 27 FCC Rcd at 10722, 10727, at ¶¶ 64, 76; AT&T-Qualcomm Order, 26 FCC Rcd at 17610-11 ¶ 49. 208 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6239 ¶285; AT&T-Leap Order, 29 FCC Rcd at 2767-68 ¶¶ 75-76; SoftBank-Sprint Order, 28 FCC Rcd at 9650, ¶ 23; Verizon Wireless-SpectrumCo Order, 27 FCC Rcd at 10724-27 ¶¶70-78; Verizon Wireless-Alltel Order, 23 FCC Rcd at 17460 ¶ 26. [check cite] For a description of some relevant competitive variables, see Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6237-6238 ¶¶ 279-280 (citing various transactions orders). 209 See, e.g., AT&T-Leap Order, 29 FCC Rcd at 2743-2744 ¶ 16; SoftBank-Sprint Order, 28 FCC Rcd at 9652 ¶ 25; Verizon Wireless-SpectrumCo Order, 27 FCC Rcd at 10711 ¶ 30; Verizon Wireless-ALLTEL Order, 23 FCC Rcd at 17462 ¶ 29; AT&T-Cingular Order, 19 FCC Rcd 21522, 21545, ¶ 43. For a general discussion of public interest conditions imposed by the FCC in certain wireless transactions granted in 2013 and 2014, see Baker, A., Brennan, T., Erb, J., Nayeem, O., Yankelevich, A., 2014. “Economics at the FCC, 2013-2014.” Mimeo. FCC, Washington, DC. (Baker, et al., (2014)). 210 See generally, AT&T-ATN Order, 28 FCC Rcd at 13670. See also, Baker, et al., (2014). 50 Federal Communications Commission DA 14-1862 spectrum licenses of Sprint and Clearwire to Softbank Corp.;212 the joint venture between GCI and ACS Wireless;213 and the assignment and lease of licenses between AT&T, Verizon Wireless, and Grain Spectrum.214 Among the smaller transactions that have occurred in the past couple of years are a number in which a nationwide provider acquired spectrum or other assets from a small or regional licensee. Not including spectrum swaps or the major transactions above, from September 2012 through June 2014 the Commission approved approximately 120 applications filed by the four nationwide providers to acquire PCS, AWS-1, Cellular, and/or 700 MHz licenses from a non-nationwide licensee – approximately 90 applications by AT&T, approximately 20 by Verizon Wireless, three by Sprint, and seven by T-Mobile. c. Additional Spectrum Initiatives 102. Another Commission initiative that could potentially make more spectrum available that would facilitate the provision of mobile wireless service is the 3.5 GHz Band proceeding. The Commission has there proposed to create a three-tier shared access authorization framework in the 3550-3650 MHz Band, with the goal of facilitating the use of small cell broadband technologies on a shared basis with incumbent federal and non- federal users of the band. The 3.5 GHz band is envisioned as an “innovation band” that would enable the exploration of new technologies and spectrum sharing with a focus on relatively low powered applications. 215 Under the proposal, access to and use of the band would be managed by a spectrum access system incorporating a geo-location enabled dynamic database.216 The three proposed tiers are: Incumbent Access, Priority Access, and General Authorized Access (GAA).217 The proposed rules would implement a framework to authorize a variety of small cell and other broadband uses of the 3.5 GHz Band via the establishment of a Citizen Broadband Radio Service to be divided into Priority Access and GAA tiers of service.218 103. On April 1, 2014, the Commission adopted a Report and Order modifying the rules governing the operation of Unlicensed National Information Infrastructure (U-NII) devices operating in the 5 GHz band.219 The new rules will make 100 megahertz of spectrum more accessible for use in homes and congested spaces like convention centers, parks, and airports and increase the potential for more unlicensed spectrum innovation. This will facilitate the provision of mobile broadband by augmenting commercial cellular networks and allowing for increased offloading to Wi-Fi networks. U-NII devices play an important role in meeting public demand for 211 See generally, AT&T-Leap Order, 29 FCC Rcd at 2735. See also, Baker, et al., (2014). 212 See generally, SoftBank-Sprint Order, 28 FCC Rcd at 9642. 213 See generally, Alaska Wireless Order, 28 FCC Rcd at 10433. See also, Baker, et al., (2014). 214 See generally, AT&T-Verizon Wireless-Grain Order, 28 FCC Rcd at 12878. 215 Amendment of the Commission’s Rules with Regard to Commercial Operations in the 3550 – 3650 MHz Band, GN Docket No. 12-354, Further Notice of Proposed Rulemaking, FCC 14-49, 29 FCC Rcd 4273(2014) (“3.5 GHz FNPRM”). The FNPRM also included a supplemental proposal to incorporate the 3650-3700 MHz band into the proposed Citizens Broadband Radio Service authorization framework. 216 Id. 217 Incumbent Access users would include authorized federal and grandfathered Fixed Satellite Service users currently operating in the 3.5 GHz Band. These users would be protected from harmful interference from Priority Access and GAA users. Priority Access Licenses would be subject to competitive bidding and would be entitled to protection from harmful interference from other Priority Access Licensees and GAA users. The GAA tier would be licensed-by-rule to permit open, flexible access to the band for the widest possible group of potential users. GAA users would have no expectation of protection from harmful interference. 218 3.5 GHz FNPRM, 29 FCC Rcd 4275, ¶ 3. 219 See, Revision of Part 15 of the Commission’s Rules to Permit Unlicensed National Information Infrastructure (U-NII) Devices in the 5 GHz Band, Report and Order, ET Docket No. 13-49, 29 FCC Rcd. 4127, (2014) 51 Federal Communications Commission DA 14-1862 wireless broadband service. Currently U-NII devices operate in 555 megahertz of spectrum in the 5 GHz band, and are used for Wi-Fi and other high-speed wireless connections. The new rules removed the restriction on indoor-only use and increase the permissible power which will provide more robust access in the 5.150-5.250 GHz band. This in turn will allow U-NII devices to better integrate with other unlicensed portions of the 5 GHz band to offer faster speeds and reduce congestion at crowded Wi-Fi hot spots such as airports and convention centers. 3. Analysis of Spectrum Holdings 104. Table IV.A.2 (Percentage Spectrum Holdings, by Provider, by Frequency Band) and Table IV.A.3 (Population-Weighted Average Megahertz Holdings by Provider, by Frequency Band) below present spectrum holdings by service provider including all spectrum bands considered suitable and available following the release of the Mobile Spectrum Holdings Report and Order.220 Table IV.A.3 shows megahertz holdings for each provider, weighted by population. Chart IV.A.1 is a graph of providers’ spectrum holdings by frequency band, measured on a MHz-POPs basis. As of June 2014, Verizon Wireless, AT&T, Sprint, and T-Mobile, together, hold close to 80 percent of all spectrum suitable and available for the provision of mobile wireless services, measured on a MHz-POPs basis.221 This is the same percentage as reported in the Sixteenth Report, but represents, in absolute terms, an increase in the total spectrum holdings of the nationwide service providers since the Sixteenth Report, given the increase in the spectrum included in the spectrum screen following the Mobile Spectrum Holdings Report and Order.222 Table IV.A.2 Percentage Spectrum Holdings, Measured on a MHz-POPs Basis by Licensee, by Frequency Band* 700 MHz Cell. SMR PCS H Block AWS-1 AWS- 4 WCS BRS EBS Spectrum 70 MHz 50 MHz 14 MHz 130 MHz 10 MHz 90 MHz 40 MHz 20 MHz 67.5 MHz 112.5 MHz*** Verizon Wireless 31.0% 48.0% 0.0% 16.2% 0.0% 38.6% 0.0% 0.0% 0.0% 0.0% AT&T 40.6% 44.6% 0.0% 29.1% 0.0% 15.8% 0.0% 91.1% 0.0% 0.0% Sprint 0.0% 0.0% 96.5% 28.0% 0.0% 0.0% 0.0% 8.6% 86.8% 69.8% T-Mobile 8.6% 0.1% 0.0% 21.7% 0.0% 40.9% 0.0% 0.0% 0.0% 0.0% US Cellular 3.6% 4.2% 0.0% 1.6% 0.0% 0.6% 0.0% 0.0% 0.0% 0.0% DISH** 6.6% 0.0% 0.0% 0.0% 100% 0.0% 100% 0.0% 0.0% 0.0% Other*** 9.7% 3.1% 3.5% 3.5% 0.0% 4.2% 0.0% 0.3% 13.2% 30.2% 220 See Mobile Spectrum Holdings Report and Order, 29 FCC Rcd at 6135 ¶ 4. 221 The holdings presented in this section include all transactions consummated as of early June, 2014, plus the assignment of AWS-1 licenses from Aloha to AT&T, which was consented to by the Commission in July 2014. See Application of AT&T Mobility Spectrum LLC and Aloha Partners II, L.P. For Consent to Assign Advanced Wireless Services A, B and C Block Licenses, ULS File No. 0006065982; Order, DA 14-1034. The holdings do not reflect other transactions consummated after June 2014, but, with one exception, these smaller transactions have minimal impacts on the numbers. The exception was the assignment of WCS A and B Block licenses from Sprint to AT&T, which represented 8.65 percent of total MHz-POPs of WCS (AT&T and Sprint Seek FCC Consent to the Assignment of WCS Licenses, 29 FCC Rcd 5879), resulting in AT&T holding essentially all of the WCS spectrum . Other smaller transactions include, but are not limited to, the exit of Cincinnati Bell and concurrent assignments and leases of spectrum to and between Grain Spectrum and Verizon Wireless (Cellco Partnership D/B/A Verizon Wireless, Cincinnati Bell Wireless, LLC, Grain Spectrum III, LLC and Grain Spectrum IV, LLC Seek FCC Consent to the Assignment of AWS, PCS, and Lower 700 MHz Band A Block Licenses, etc., 29 FCC Rcd 5368), the assignment of 700 MHz A-Block spectrum from Actel to T-Mobile (ULS File No. 0006402872). 222 Sixteenth Report, 28 FCC Rcd at 3787 ¶ 118. 52 Federal Communications Commission DA 14-1862 * Estimates include all transactions consummated as of the beginning of June, 2014. Abbreviations for spectrum bands: Cell. (Cellular); SMR (Specialized Mobile Radio Service), BRS (Broadband radio Service), EBS (Educational Broadband service). ** Dish Network Corporation currently does not provide mobile service. *** In the application of the spectrum screen in secondary market transactions, 89 megahertz of EBS spectrum is included. Table IV.A.3 Population-Weighted Average Megahertz Holdings by Licensee, by Frequency Band* 700 MHz Cell. SMR PCS H Block AWS- 1 AWS- 4 WCS BRS EBS Spectrum Counted 70 MHz 50 MHz 14 MHz 130 MHz 10 MHz 90 MHz 40 MHz 20 MHz 67.5 MHz 112.5 MHz*** Verizon Wireless 21.7 24.6 0.0 21.1 0.0 34.7 0.0 0.0 0.0 0.0 AT&T 28.4 22.8 0.0 38.0 0.0 14.2 0.0 18.3 0.0 0.0 Sprint 0.0 0.0 13.9 36.6 0.0 0.0 0.0 1.7 58.6 78.5 T-Mobile 6.0 0.0 0.0 28.3 0.0 36.8 0.0 0.0 0.0 0.0 US Cellular 2.5 2.2 0.0 2.1 0.0 0.5 0.0 0.0 0.0 0.0 DISH** 4.6 0.0 0.0 0.0 10.0 0.0 40.0 0.0 0.0 0.0 Other*** 6.8 1.6 0.5 4.5 0.0 3.8 0.0 0.1 8.9 34.0 * Estimates in Table IV.A.3 include all transactions consummated as of the beginning of June, 2014. ** Dish Network Corporation currently does not provide mobile service. *** In the application of the spectrum screen in secondary market transactions, 89 megahertz of EBS spectrum is included. Source: FCC staff estimates generally based on transactions consummated as of the beginning of June, 2014. 105. Chart IV.A.1 above shows the population-weighted spectrum holdings of nationwide wireless providers by frequency. It provides a side-by-side comparison of each licensee’s total spectrum holdings by band, measured by population-weighted average megahertz. We consider population-weighted spectrum holdings in order to account for the variation of customer bases in different geographic areas. A spectrum license in Los - 10 20 30 40 50 60 70 Verizon Wireless ATT Sprint T-Mobile US Cellular DISH Other M H z- P op s (B ill io ns ) Chart IV.A.1 Mobile Wireless Provider Spectrum Holdings by Band Weighted by Population June 2014 WCS EBS BRS AWS-4 AWS H Block PCS SMR Cellular 700 MHz 53 Federal Communications Commission DA 14-1862 Angeles or New York City, for example, covers more customers than a spectrum license over the same amount of land area in White Sands, NM. 106. Below-1-GHz spectrum includes Cellular (850 MHz), SMR (800/900 MHz), and the 700 MHz band. Of this spectrum, Verizon Wireless and AT&T each hold a significant amount of the available Cellular and 700 MHz spectrum. In particular, when measured on a licensed MHz-POP basis, Verizon Wireless holds approximately 38 percent of the licensed MHz-POPs of the combined Cellular and 700 MHz band spectrum, while AT&T holds approximately 42 percent. Additionally, T-Mobile holds approximately five percent of these bands and US Cellular holds approximately four percent. A number of other smaller licensees, combined, hold the remaining approximately 11 percent of the Cellular and 700 MHz band spectrum. Sprint holds 96.5 percent of the SMR spectrum. Providers also vary with respect to their below-1-GHz spectrum holdings according to population density as seen in Chart IV.A. 2 below. In particular, AT&T and T-Mobile have focused their low frequency spectrum acquisition on urban centers, Sprint and Verizon Wireless have purchased their licenses in both urban and rural areas, and the other smaller providers hold more spectrum in rural areas than they do in urban areas. Source: FCC staff estimates generally based on transactions consummated as of the beginning of June, 2014. 107. All four nationwide providers hold substantial amounts of above-1-GHz spectrum. Verizon Wireless, AT&T, and T-Mobile each hold a substantial number of PCS and AWS-1 spectrum licenses, while Sprint holds significant amounts of PCS spectrum. In the PCS and AWS-1 spectrum bands, no licensee holds more than 41 percent of the total MHz-POPs for either of these two bands. Verizon Wireless holds approximately 25 percent of the licensed MHz-POPs of the combined PCS and AWS-1 band spectrum, AT&T holds approximately 24 percent, Sprint holds approximately 17 percent, and T-Mobile holds approximately 30 percent. Regional provider US Cellular holds approximately one percent of the combined PCS and AWS-1 band spectrum, Verizon Wireless AT&T Sprint Nextel T-Mobile US Cellular DISH C-Spire 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 1st Decile 2nd Decile 3rd Decile 4th Decile 5th Decile 6th Decile 7th Decile 8th Decile 9th Decile 10th Decile 0.1% 0.4% 0.5% 0.7% 1.3% 2.1% 3.1% 5.8% 11.5% 74.4% A ve ra ge S ub 1 G H z Sp ec tr um (P op - W ei gh te d M H z) Chart IV.A.2 Average Below-1-GHz Spectrum by Population Density Deciles US Population Deciles Sorted by County Population Density Verizon Wireless AT&T Sprint Nextel T-Mobile US Cellular DISH C-Spire Other % of Pops % of Area More Dense Less Dense Other 54 Federal Communications Commission DA 14-1862 while other smaller providers hold the remaining four percent. In addition to its PCS and AWS-1 holdings, AT&T holds approximately 91 percent of the licensed MHz-POPs of WCS spectrum.223 Sprint holds a predominant amount of 2.5 GHz spectrum, comprised of the BRS and EBS bands, the highest frequencies currently considered suitable and available for the provision of mobile broadband service.224 B. Non-Spectrum Input Segments 1. Wireless Infrastructure 108. Wireless infrastructure facilities hosting cellular base stations are a major input into the provision of mobile wireless services. They include towers and other tall structures for macro sites, such as lattice towers, guyed towers, monopoles, rooftops, water towers, and steeples. In addition to the use of towers and other tall structures, wireless infrastructure also include distributed antenna systems (DAS)225 and facilities for small cell technologies226 that are generally deployed to address coverage and capacity issues indoors, in densely populated areas outdoors, and even underground.227 For example, small cells and DAS antennas can be placed on utility poles, buildings, or traffic signal poles, in areas where constructing towers is not feasible or wireless traffic demands are too great to be met solely with fewer, large cells.228 In order to expand capacity and improve coverage, wireless service providers are also relying more on Heterogeneous Networks (“HetNets”) that use a mix of traditional macro cells, DAS, and small cells, to tailor coverage and capacity to best serve a particular location.229 109. The number of cell sites of all types in use by providers continues to grow in order to satisfy the increased demand for mobile wireless services, to expand geographic service area coverage, to improve coverage in existing service areas, and to accommodate newer technologies. Large-scale deployments of small cells are only starting to take off in the United States.230 According to CTIA, there were 304,360 cell sites in use at year- 223 See AT&T-WCS Order, 27 FCC Rcd at 16461-16462 ¶ 4-6; see generally, WCS Order on Reconsideration. 224 See SoftBank-Sprint Order, 28 FCC Rcd at 9645 ¶ 11. 225 A DAS is a network of antennas typically connected by fiber optic cables to a central hub housing transceiver that is linked to macrocellular network. See The DAS Forum, Distributed Antenna Systems (DAS) and Small Cell Tech ologies Distinguished, February 4, 2013, at 2, available at http://www.thedasforum.org/wp-content/uploads/2013/02/DAS-And- Small-Cell-Technologies-Distinguished-2_4_13.pdf (visited July 25, 2014). 226 A small cell normally refers to an operator-controlled, low-powered radio access node deployed at a particular location that include both antenna and transceiver operating in licensed spectrum or unlicensed carrier-grade Wi-Fi. Small cells typically have a range from 10 meters (or 11 yards) to several hundred meters (or yards). Types of small cells include femtocells, picocells, metrocells and microcells – broadly increasing in size from femtocells (the smallest) to microcells (the largest). See Small Cell Forum, What is a Small Cell, at http://www.smallcellforum.org/aboutsmallcells-small-cells-what-is- a-small-cell (visited July 25, 2014). 227 See HetNet Forum, DAS & Small Cell Solutions: Improving In-Building Wireless, at 4, available at http://www.thedasforum.org/wp-content/uploads/2013/07/HITEC-presentation-from-HetNet-Forum-final.pdf, and (visited July 25, 2014). See also Tammy Parker, All Four National Mobile Carriers to Use NYC Subway DAS, FIERCEWIRELESS, Apr. 28, 2013. 228 Because DAS sites are less visible than tower structures, they may be particularly desirable in areas with stringent siting regulations, such as historic districts. See Sixteenth Competition Report at ¶ 318. 229 HetNets are an emerging deployment option for service providers. See 4G AMERICAS, Developing and Integrating a High Performance Het-Net 2, October 2012. See also PCIA Comments at fn. 3 (“‘Heterogeneous network’ is a term used to describe the combination of ‘macro’, or large, infrastructure such as monopoles with small cells and distributed antenna systems. By integrating the two types of infrastructure together, providers are able to target geographic areas to increase network capacity.”). 230 See Phil Goldstein, Crown Castle Sees Small Cell Opportunity as Big as Cell Towers in Early 2000s, July 25, 2014, http://www.fiercewireless.com/story/crown-castle-sees-small-cell-opportunity-big-cell-towers-early-2000s/2014-07-25 (visited July 29, 2014). See also, FierceWirelessTech, AT&T’s Mansfield: Outdoor LTE Small Cell Market will Ramp Up 55 Federal Communications Commission DA 14-1862 end 2013.231 This represents a 0.9 percent (or 2,581) increase in the number of cell sites from the year-end 2012, a 26 percent increase in the past five years (since December 31, 2008), and an 87 percent increase in the past ten years (since December 31, 2003).232 110. A specialized communications tower industry has developed to provide and manage support structures for the cell sites required by mobile wireless service providers by leasing space to them. Today, there are more than 80 tower and DAS operators in the United States,233 and a majority of towers are now owned or operated by independent companies rather than mobile wireless service providers.234 Independent tower operators own, operate and lease shared wireless communications and broadcasting towers, manage other high structure sites (such as rooftops, water towers) for property owners, and to a lesser extent, build and operate DAS networks and small cell facilities for mobile service providers.235 In most cases, tower operators and property owners have an incentive to increase their business by leasing antenna, rooftop and other site space to as many wireless service providers as possible.236 According to PCIA, the three largest publicly-traded neutral host providers – Crown Castle, American Tower and SBA Communications – own and operate more than 66,000 towers.237 Another estimate indicates that these three tower operators own, manage, or operate more than 83,000 towers in the United States out of a total 102,000 as of January 2014.238 The availability of leased space on existing towers may eliminate the need to build new towers for competing service providers and new entrants in a market, reduce the capital requirements for network deployments and capacity expansion, and facilitate entry of new wireless service Next Year, June 24, 2014, available at http://www.fiercewireless.com/tech/special-reports/atts-mansfield-outdoor-lte-small- cell-market-will-ramp-next-year (visited July 29, 2014). 231 See CTIA, 2013 Annual Wireless Industry Survey Results (“CTIA2013”), at 114, June 2014. Because multiple cell sites can be co-located in the same “tower” site, the reported cell sites should not be equated with “towers.” See also CTIA2013 at 105. The reported cell sites include repeaters and other cell-extending devices (e.g., femtocells, or distributed antenna systems). See CTIA2013 at 105 and 106. 232 The incremental cell site count might be skewed by changes in the survey population, and not reflect actual new cell site deployment. See CTIA2013 at 109. All calculations based on CTIA2013 Table 35 at 107. Appendix Table IV.B.i provides further breakout details on the number of reported cell sites per provider. Cell site counts for individual service provider are from US Wireless 411: Version 51, Figure 51 at 30, March 2014. The total industrywide cell count is from CTIA2013, at 114, June 2014. 233 See http://www.wirelessestimator.com/t_content.cfm?pagename=US-Cell-Tower-Companies-Complete-List (list of tower operators). 234 Some major wireless service providers have sold or in the process of selling their tower business to third party tower operators. See AT&T News Release, AT&T and Crown Castle Close $4.83 Billion Tower Transaction, December 16, 2013; Crown Castle News Release, Crown Castle Completes Tower Transaction With T-Mobile USA, November 30, 2012;Sprint Nextel News Release, Sprint Nextel Completes Tower Sale to TowerCo for Approximately $670 Million in Cash, September 24, 2008. 235 See American Tower 2013 Annual Report at 1, Crown Castle 2013 Annual Report (10-K) at 1, SBA Communications 2013 Annual Report (10-K) at 1. 236 See American Tower, Investor Relations (“Our primary business is leasing antenna space on multiple-tenant communications sites to wireless service providers, radio and television broadcast companies, wireless data providers, government agencies and municipalities and tenants in a number of other industries”), at http://www.americantower.com/corporateus/investor-relations/index.htm. See also Verizon Network Real Estate Inquires (“Verizon Wireless receives thousands of inquiries each year from property owners, property managers and customers who offer property on which our communications facilities can be located”), at http://www.verizonwireless.com/b2c/realestate/. 237 See PCIA Comments at 8 238 See http://www.wirelessestimator.com/t_content.cfm?pagename=US-Cell-Tower-Companies-Complete-List (list of tower operators, with Crown Castle 39,739, American Tower 28,463, and SBA Communications 14,873 as of January 2014. Not including DAS structures and rooftops. Visited July 25, 2014). 56 Federal Communications Commission DA 14-1862 providers into a market. 111. Based on tower site information (including towers, rooftops, and DAS) collected by Commission staff from eleven large tower operators (owning or operating close to 90 percent of towers in the United States),239 as of September 2013, 89 percent of counties have more than three tower operators, and 50 percent have more than six tower operators (see Chart IV.B.1). Note: Data based on eleven tower companies. These companies are Crown Castle, American Tower, SBA Communications, KGI Wireless, Global Tower Partners, AT&T Towers, T-Mobile Towers, InSite Towers, SubCarrier Communications, Clear Channel, Central States Tower, CTI Towers, Skyway Towers, and Pegasus Wireless. Population density is from 2010 census. 112. Based on the data, tower operators build and operate more towers and DAS nodes in densely populated areas in order to support better coverage and more wireless data usage. For example as of September 2013, the average number of tower and DAS sites per county is 29 for counties with a population density between 75 and 100 persons per square mile, compared with an average of 377 per county for counties with a population density between 2000 and 4000 (see Chart IV.B.2). 239 They are Crown Castle, American Tower, SBA Communications, KGI Wireless, Global Tower Partners, AT&T Towers, T-Mobile Towers, InSite Towers, SubCarrier Communications, Clear Channel, Central States Tower, CTI Towers, Skyway Towers, and Pegasus Wireless. Tower and cell site information is downloaded in the first week of September, 2014 from their websites that can be found at http://www.wirelessestimator.com/t_content.cfm?pagename=US-Cell-Tower-Companies- Complete-List. 128 222 323 384 510 524 446 304 203 113 38 0 100 200 300 400 500 600 1 2 3 4 5 6 7 8 9 10 11N um be r of C ou nt ie s Number of Site Operators in a County Chart IV.B.1 Number of Counties with a Certain Number of Site Operators 57 Federal Communications Commission DA 14-1862 Note: The number of sites has been rounded to the nearest decimal. Data based on eleven tower companies referenced above. Counties considered rural are those with fewer than 100 persons per square mile. 113. In addition, there are also more tower operators in densely populated counties (often associated with smaller land areas) than less populated counties (often associated with larger land areas). The numbers range from two operators per county in rural counties with one person or less per square mile and an average land size of 11,122 square miles to more than seven operators in dense urban counties with a population density of more than 4000 and an average land size of 98 square miles (see Chart IV.B.3). 240 240 Population density is from the 2010 Census. 6 8 14 18 25 29 47 52 83 102 135 120 220 377 382 0.0 100.0 200.0 300.0 400.0 500.0 A ve S it e C nt s pe r C nt y County Population Density Range (persons per square mile) Chart IV.B.2 Average Tower and DAS Counts per County by Population Density 2014 Rural 58 Federal Communications Commission DA 14-1862 Note: Data based on eleven tower companies referenced above. Counties considered rural are those with fewer than 100 persons per square mile. 114. Two significant constraints faced by wireless infrastructure providers that need to add or modify tower and DAS sites are capital expenditure, and the need to obtain necessary regulatory and zoning approvals from local and federal authorities.241 Below, we briefly discuss each of these constraints. In terms of capital expenditure, collocating wireless equipment on existing structures is often the most efficient and economical solution for mobile wireless service providers that need new cell sites, either to expand their existing coverage area, increase their capacity, or deploy 4G broadband services. The average cost to build a new tower is between $250,000 and $300,000, whereas the average cost of collocation on an existing tower is less than 25 percent of the total cost of a new tower.242 The largest, publicly-traded infrastructure companies alone made capital expenditures of approximately $1.16 billion in 2012, up from $1.02 billion in 2011.243 Their capital expenditures for 2013 are estimated to have been between $1.3 and $1.5 billion. Significant portions of these capital expenditures are dedicated to network improvement, including new site construction and improvements to existing sites to accommodate more provider facilities. In 2012, these network improvement expenditures totaled approximately $658 million, compared to $306 million in 2011. Such investments are estimated to have been up to $917 million 241 Delays in the zoning approval process were the subject of a Petition for Declaratory Ruling filed by CTIA in 2008. On November 18, 2009, the Commission adopted a Declaratory Ruling which, among other things, defined presumptively reasonable time parameters for state or local zoning authorities to decide whether or not to approve a cell site application. See Sixteenth Report at ¶ 326. There is no evidence that shortages of transmission equipment, including antennas, to install at cell sites act as a barrier to cell site deployment. See Sixteenth Competition Report, ¶ 325. 242 See PCIA Comments at 8. See also Martha DeGrasse, AT&T Cell Site of the Future Hits a Speed Bump, July 17, 2014 (“the cost per site was coming in at $380,000” in New York metro area), available at http://www.rcrwireless.com/article/20140717/infrastructure-2/att-cell-site-future-hits-speed-bump/ (visited July 28, 2014). 243 See PCIA Comments at 6. 98 340 724 598 1,027 0 2 4 6 8 < = 1 (1 , 1 0] (1 0, 2 5] (2 5, 5 0] (5 0, 7 5] (7 5, 1 00 ] (1 00 , 1 50 ] (1 50 , 2 00 ] (2 00 , 2 50 ] (2 50 , 5 00 ] (5 00 , 7 50 ] (7 50 , 1 00 0] (1 00 0, 2 00 0] (2 00 0, 4 00 0] > 40 00 7.7 8.0 7.6 6.6 7.5 7.8 7.6 6.9 6.7 6.2 6.0 5.3 4.5 3.3 2.1 A ve L an d A re a pe r C ou nt y (s qu ar e m il e A ve T ow er O pt r C ou nt s pe r C ou nt y County Population Density Range (persons per square mile) Chart IV.B.3 Ave Tower Operator Counts per County by Pop Density and County Land Area 2013 Urban Rural 59 Federal Communications Commission DA 14-1862 in 2013.244 The total annual expenditure for structures by wireless service providers (excluding satellite service providers) was estimated to be $5,741 billion for 2012, almost 45 percent increase from $3,966 Billion in 2011, or 48 percent increase from $3,890 billion in 2009.245 115. Recent FCC Initiatives. Section 6409 of the Middle Class Tax Relief and Job Creation Act of 2012 provides that a state or local government “may not deny, and shall approve” any request for collocation, removal, or replacement of transmission equipment on an existing wireless tower or base station, provided this action does not substantially change the physical dimensions of the tower or base station.246 In January 2013, the Commission offered its interpretive guidance to assist parties in understanding their obligations under Section 6409(a).247 In August 2014, the Commission adopted a Report and Order to streamline and eliminate outdated provisions of the Part 17 Rules governing the construction, marking and lighting of antenna structures. 248 In 2013, the Commission began a rulemaking proceeding to consider certain options to reduce regulatory barriers and streamline process at the local level and at the Commission for infrastructure deployment.249 On October 21, 2014, the Commission adopted an order that eliminated unnecessary reviews, and therefore costs and delays, for wireless facilities siting.250 The Commission has also entered into two Nationwide Programmatic Agreements (NPA) with the Advisory Council on Historic Preservation (ACHP) and the National Conference of State Historic Preservation Officers (NCSHPO) – (1) to clarify the National Historic Preservation Act (NHPA) Section 106 process for new tower construction,251 and (2) to provide guidance on collocation of communications equipment on existing tower structures.252In addition, the Commission has taken steps with relevant government and non-governmental stakeholders to develop a process for “clearing” existing towers that were not subject to historic preservation review prior to construction, including “twilight towers that were not required to, and did not, complete the Section 106 historic preservation review process.”253 Once complete, this effort will make thousands of additional towers available for collocation.254 244 See PCIA Comments at 6. 245 See CTIA2013 Chart 18 at 112 (citing information from the United States Census). 246 Middle Class Tax Relief and Job Creation Act of 2012, Pub. L. No. 112-96, § 6409(a) (2012). 247 Wireless Telecommunications Bureau Offers Guidance on Interpretation of Section 6409(a) of the Middle Class Tax Relief and Job Creation Act of 2012, Public Notice, 28 FCC Rcd. 1 (Jan. 25, 2012), http://hraunfoss.fcc.gov/edocs_public/attachmatch/DA-12-2047A1_Rcd.pdf. 248 See FCC News Release, FCC Streamlines Part 17 Rules to Provide Clarity Regarding Antenna Structure Lighting and Marking, August 8, 2014. 249 See Acceleration of Broadband Deployment by Improving Wireless Facilities Siting Policies; Acceleration of Broadband Deployment: Expanding the Reach and Reducing the Cost of Broadband Deployment by Improving Policies Regarding Public Rights of Way and Wireless Facilities Siting; 2012 Biennial Review of Telecommunications Regulations, WT Docket No. 13-238, WC Docket No. 11-59, WT Docket No. 13-32, Notice of Proposed Rulemaking, 28 FCC Rcd 14238 (2013). 250 See Acceleration of Broadband Deployment by Improving Wireless Facilities Siting Policies; Acceleration of Broadband Deployment: Expanding the Reach and Reducing the Cost of Broadband Deployment by Improving Policies Regarding Public Rights of Way and Wireless Facilities Siting; 2012 Biennial Review of Telecommunications Regulations, WT Docket Nos. 13-238, 11-59, 13-32, Report and Order, FCC 14-143 (rel. Orc. 21, 2104). 251 See 47 C.F.R. 1, Appendix C 252 Id at appendix B. 253 See Section 106 of the National Historic Preservation Act (NHPA) 16 U.S.C. 470(f). "Twilight Towers" are towers built between March 16, 2001 and March 7, 2005. 254 Wireless Infrastructure Report and Order at 4. . 60 Federal Communications Commission DA 14-1862 2. Backhaul 116. Backhaul facilities link a mobile wireless service provider’s cell sites to the mobile switching centers that provide connections to the provider’s core network, the public switched telephone network, or the Internet, carrying wireless voice and data traffic for routing and onward transmission. Backhaul connections are an integral component of a wireless service provider’s network, and the cost of backhaul is approximately 30 percent of the operating cost of providing wireless service.255 Backhaul services are generally provided by incumbent local exchange carriers (ILECs); competitive local exchange carriers (CLECs); competitive fiber and microwave wholesalers; cable providers; and independent backhaul operators.256 In some areas, the vast majority of existing wireless towers have fiber connections.257 High bandwidth fiber backhaul also allows wireless service providers to rapidly deploy 4G LTE services,258 while a lack of fiber backhaul can delay a provider’s LTE rollout.259 117. Mobile backhaul needs will keep increasing as wireless providers continue to deploy LTE technology in their networks and mobile subscribers use mobile devices for more data intensive applications, such as mobile video streaming.260 Analysts project that demand for mobile backhaul will grow by 9.7 times between 2011 and 2016.261 Infonetics Research estimates that the annual investment in the backhaul market will be in the range of $8 to $9 billion over the next few years262 even with a slower projected growth in the demand for mobile backhaul.263 This is a significant increase given that the entire market was worth less than $5 billion in 2009.264 Infonetics Research projects that telecom service providers will collectively spend $43 billion on backhaul over five years from 2013 – 2017.265 255 See Sprint Comments at 6 256FierceTelecom.com, Telco Backhaul Strategies, at 1-2, November 2011. Providers of backhaul services include ILECs such as AT&T, Verizon, and CenturyLink; CLECs such as Level 3, tw telecom inc., Cbeyond, Inc., and XO Communications, LLC; competitive fiber and microwave wholesalers such as Level 3, FPL FiberNet, IP Networks, and Zayo; cable providers such as Charter Communications, Comcast Business, Cox Carrier Services, and Time Warner Cable Business Class; independent backhaul operators, including backhaul specialists such as Telecom Transport Management, and Tower Cloud, and potentially some tower operators 257 See Windstrem Annual Report, 2013, page F-3. 258 See SASCHA SEGAN, Exclusive: T-Mobile CTO Talks Carrier's Journey to Fastest LTE Network, January 8, 2014, available at http://www.pcmag.com/article2/0,2817,2429285,00.asp (last visited April 3, 2014). 259 See Phil Goldstein, Sprint's LTE rollout hampered by lack of backhaul and Network Vision issues, July 24, 2013, available at http://www.fiercewireless.com/story/sprints-lte-rollout-hampered-lack-backhaul-and-network-vision-issues/2013-07-24. 260 Clayton Funk, et. al. Trends and forecasts for the Wireless tower Industries, September 2013 at 50. http://www.aaeconline.com/wp-content/uploads/2013/11/AGL-Trends-and-Forecasts-9-2013.pdf. 261 See Verizon Wireless Comment at 58 citing FIERCEWIRELESS U.S. Mobile Backhaul Demand Forecast to Grow More Than Nine Times in the Next Four Years (Mar. 13, 2012), http://www.fiercemobilecontent.com/press-releases/us-mobile- backhaul-demand-forecast-grow-more-ninetimes-next-four-years. (The global demand for mobile backhaul equipment is projected to reach $10.4 billion in 2014 (compared to $7.2 billion in 2009) and Infonetics Research Press Release, Shift Seen in Operator Strategy for Mobile Backhaul; Equipment Spending Up 21% (Apr. 21, 2010), http://www.infonetics.com/pr/2010/Mobile-Backhaul-and-Microwave-Market-Highlights.asp. 262 Infonetics Research, Mobile backhaul market passes $8 billion, driven by HSPA/HSPA+ and LTE deployments, April 2, 2013 http://www.infonetics.com/pr/2013/2H12-Macrocell-Mobile-Backhaul-Market-Highlights.asp. 263 Infonetics Research, Macrocell Mobile Backhaul Market Growth Decelerating, October 30, 2013. http://www.marketwatch.com/story/infonetics-research-macrocell-mobile-backhaul-market-growth-decelerating-2013-10-30. 264 Infonetics Research, Mobile backhaul market passes $8 billion, driven by HSPA/HSPA+ and LTE deployments, April 2, 2013 265 Infonetics Research, Macrocell Mobile Backhaul Market Growth Decelerating, October 30, 2013. 61 Federal Communications Commission DA 14-1862 118. As the mobile data traffic has grown rapidly in recent years, major mobile wireless providers have deployed or are in the process of deploying Ethernet backhaul either over fiber or microwave to their cell sites. Over 90 percent of AT&T’s data traffic is on enhanced backhaul as of June 2013.266 T-Mobile already has upgraded backhaul facilities for its 40,000 LTE sites as of March 2014 and is expected to update the remaining 15,000 sites by the summer of 2015.267 In connection with its Network Vision Plan, Sprint has deployed 33,000 LTE/Network Vision sites with Ethernet backhaul covering more than 200 million people at the end of 2013 and is expected to cover 250 million people with its LTE sites by the middle of 2014.268 Verizon Wireless also deployed fiber backhaul facilities for its 4G LTE sites,269 which carries 69 percent of its data traffic as of the fourth quarter of 2013.270 119. Recent FCC Initiatives. The Commission has examined issues related to backhaul including special access services and the use of microwave spectrum for backhaul services.271 In January 2005, the Commission started a proceeding to broadly examine the regulatory framework for local exchange carriers’ (LECs) interstate special access services.272 On August 22, 2012 the Commission adopted a Report and Order that suspended, on an interim basis, rules that allowed for automatic grants of pricing flexibility for special access services in light of evidence in the record the rules failed to accurately reflect the state of competition in the market for special access.273 On September 15, 2014 the Commission’s Wireline Competition Bureau issued an Order on Reconsideration that set a deadline of December 15, 2014 for providers to submit data on networks, prices, and terms for special access in order to enable the Commission to assess the state of competition.274 In August 2013, the Commission significantly modified the Commission's Part 15 rules governing unlicensed communication equipment in the 57-64 GHz band to enhance the use of unlicensed spectrum as a relatively low- cost, high-capacity short-range backhaul alternative to connect wireless broadband networks and for other wireless applications.275 266 See AT&T 4G LTE Network Ranked Fastest For Second Straight Year, AT&T news release, June 10, 2013. 267 See Neal Gompa, “T-Mobile’s LTE will cover 250 million people in 2014, everywhere in the US by 2015,” March 14, 2014, available at http://www.extremetech.com/electronics/178517-t-mobiles-lte-will-cover-250-million-people-in-2014- everywhere-in-the-us-by-2015 (visited April 9, 2014). See also Sascha Segan, “Exclusive: T-Mobile CTO Talks Carrier's Journey to Fastest LTE Network,” January 8, 2014, available at http://www.pcmag.com/article2/0,2817,2429285,00.asp (visited September 18, 2014). 268 See Sprint Quarterly Presentation for the 4th Quarter 2013, Feb. 11, 2014, at 14, and BENNY HAR-EVEN, Interview: Senior Systems Engineer, Sprint: “We expect that small cells will be key to 2500MHz network densification,” October 20, 2013, available at http://lteconference.wordpress.com/2013/10/30/senior-systems-engineer-sprint-we-expect-that-small-cells- will-be-key-to-2500mhz-network-densification/. 269 See Transcript for Verizon at Oppenheimer Holdings Inc Technology, Internet & Communications Conference, at 7, August 15, 2012, available at http://www.verizon.com/investor/DocServlet?doc=oppenheimer_vz_transcript.pdf. 270 See Verizon presentation for the 4th Quarter 2013 earnings, at 10, available at http://www.verizon.com/investor/qreport_4q_2013_quarter_earnings_01212014.htm. 271 See Sixteenth Competition Report at ¶ 336 – 338. 272 See 2005 Special Access NPRM. 273 See, In the Matter of Special Access for Price Cap Local Exchange Carriers; WC Docket No. 05-25 and AT&T Corporation Petition for Rulemaking to Reform Regulation of Incumbent Local Exchange Carrier Rates for Interstate Special Access Services. RM-10593, 27 FCC Rcd 10557 (2012). 274 See Special Access for Price Cap Local Exchange Carriers; AT&T Corporation Petition for Rulemaking to Reform Regulation of Incumbent Local Exchange Carrier Rates for Interstate Special Access Services, WC Docket No. 05-25, RM- 10593, Order on Reconsideration, DA 14-1327 (Wireline Comp. Bur. rel. Sept. 15, 2014) (Reconsideration Order) 275 See FCC new release FCC MODIFIES PART 15 RULES TO SPUR THE DEPLOYMENT OF WIRELESS SERVICES, UNLICENSED SPECTRUM INNOVATION IN THE 57-64 GHZ BAND, August 9, 2013. 62 Federal Communications Commission DA 14-1862 63 Federal Communications Commission DA 14-1862 C. Downstream Mobile Wireless Ecosystem 120. Handsets and devices are a central part of consumers’ mobile wireless experience. In general, smartphone adoption has increased significantly over the past few years. In 2011, 46 percent of mobile wireless consumers reported to have smartphones, and 60 percent of consumers who purchased a new mobile device during the fourth quarter selected a smartphone over a feature phone.276 By January 2014, 68 percent of mobile wireless consumers reported using smartphones, and 84 percent of new phone purchases were smartphones.277 Since Apple entered the smartphone business with the iPhone in June 2007, many handset manufacturers have introduced competing products with similar features such as touch screens, mobile web browsing capabilities, and current-generation operating systems. During 2011, the iPhone exclusive handset arrangement between Apple and AT&T ended, and multiple service providers began offering the iPhone on their networks.278 In 2012, Verizon Wireless, Sprint and other providers started selling the iPhone 4s and iPhone 5s. Innovative smartphones that are not subject to exclusive arrangements are widely available. Popular smartphone operating systems such as the Android and the Apple iOS are available from multiple service providers, permitting consumers to pair their preferred operating systems with different service providers. 121. The operating system of a smartphone is one of the major factors that determine the smartphone’s ability to support mobile applications and Internet-based services. Apple’s iOS and Google’s Android have emerged as the two leading mobile operating systems.279 According to ComScore, by August 2014, 174 million people in the US owned smartphones. Android’s share of the smartphone operating system grew from three percent in May 2009 to 52 percent inAugust 2014, while iOS’s market share increased from 20 percent to 42 percent over the same period.280 Over essentially the same period, (Research in Motion) RIM’s market share has declined from the top position to one of less than two percent of the market.281 122. Until recently, the prevailing model for the distribution of handsets to U.S. postpaid mobile subscribers was the handset subsidy model, where consumers were offered a bundling contract in which a provider conditioned the sale of a subsidized handset upon the consumer’s agreement to purchase a multi-month wireless service subscription, typically for a minimum of one or two years in a postpaid service plan282 with a locked handset that could only be used with the particular provider As discussed in more detail below, service providers increasingly are offering equipment installment payment (EIP) plans as an alternative to traditional handset subsidies.283 Under these plans, consumers purchase the device at the full price, but instead of paying the cost upfront, are billed in monthly installment payments. Many service providers generally unlock phones at the customer’s request when the service terms have been fulfilled, subject to a certain number of conditions, e.g., a limit on the number of devices that can be unlocked, a minimum number of days of activation, and that the device 276 Sixteenth Report, 28 FCC Rcd at 3836, ¶ 339. 277 Nielsenwire, The Nielsen Company, Multiplying Mobile: How Multicultural Consumers are Leading Smartphone Adoption, March 4, 2014. 278 Prior to 2011, Apple distributed its iPhone through AT&T (and its affiliates) only. An exclusive handset arrangement (EHA) is an arrangement in which a handset manufacturer or vendor agrees to sell a particular handset model to only one wireless service provider, usually for a specified period of time. See Fifteenth Report, 26 FCC Rcd at 9853 ¶ 332. 279 See ComScore press release, ComScore Reports August 2014 US Subscriber Market Share. 280 ComScore press release, ComScore Reports August 2014 US Subscriber Market Share. 281 ComScore press release, ComScore Reports August 2014 US Subscriber Market Share. 282 See Antitrust Law and Economics, at 326 (“Under a tying arrangement, the seller of a product conditions the sale of one product upon the buyer’s agreement to purchase a second product.”) In particular, the sale of the handset is conditioned on the subsequent purchase of the multi-month wireless service subscription. 283 See Section V.A infra for a detailed discussion. 64 Federal Communications Commission DA 14-1862 has not been reported lost or stolen.284 123. Cell Phone Unlocking. On February 11, 2014, CTIA-The Wireless Association adopted six principles on cell phone unlocking into their evolving Consumer Code for Wireless Service (“Consumer Code”).285 Under these six principles, wireless providers will disclose their unlocking policies, unlock postpaid devices when contracts have been fulfilled, unlock prepaid devices within a year of activation, notify customers of eligibility for unlocking at the time of eligibility, unlock devices with a reasonable period of time, and unlock devices for deployed personnel.286 The Consumer Code included a three month timeline for implementation of half of the principles and a one year timeline for implementation of all of the principles by February 11, 2015.287 124. Nationwide wireless providers have begun implementation of the unlocking policies in accordance with the timeline in the Consumer Code. Verizon, AT&T, Sprint, and T-Mobile have established dedicated webpages disclosing their unlocking policies. AT&T has created an online unlocking portal to enable online unlocking requests.288 Sprint and T-Mobile provide toll free customer service numbers to call for unlocking codes.289 Verizon does not generally lock its devices.290 Verizon’s Phone-in-the-Box prepaid service is restricted to Verizon’s prepaid service for six months and Verizon’s network for 12 months.291 After six months, Verizon automatically and remotely removes the restriction on these devices to the prepaid network.292 In August 2014, T-Mobile deployed an unlocking application that allows customers to request unlocking on a permanent or temporary basis for travel.293 Some MVNOs that have committed to comply with the Consumer Code, like Tracfone,294 have also begun implementation of the unlocking principles in the Consumer Code.295 125. A handset that functions on one network may not be compatible with a network using the same air interface technology if the networks operate on different spectrum bands. For example, T-Mobile’s WCDMA handsets operate in the AWS-1 spectrum (1.7/2.1 GHz band) while AT&T’s WCDMA handsets operate in the 284 See Section II.C.2, infra and Verizon Wireless, Customer Agreement, available at http://www.verizonwireless.com/b2c/globalText?textName=CUSTOMER_AGREEMENT&jspName=footer/customerAgree ment.jsp (last visited April 11, 2014); Sprint, Standardized Wireless Service Checklist, available at http://www.sprint.com/landings/ctiachecklist/docs/ctia-transparency-postpaid.pdf (last visited April 11, 2014); and T- Mobile Support blog post, “Unlock Your Phone with a SIM Unlock Code,” updated on March 11, 2014, available at http://support.t-mobile.com/docs/DOC-1588 (last visited April 11, 2014). 285 https://www.fcc.gov/device-unlocking-faq; see also https://fcc.github.io/device-unlocking/ 286 http://www.ctia.org/policy-initiatives/voluntary-guidelines/consumer-code-for-wireless-service 287 http://www.ctia.org/policy-initiatives/voluntary-guidelines/consumer-code-for-wireless-service 288 https://www.att.com/deviceunlock/client/en_US/ 289 https://www.sprint.com/legal/unlocking_policy.html; http://support.t-mobile.com/docs/DOC-1588 290 https://www.verizonwireless.com/aboutus/commitment/safety-security/device-unlocking-policy.html 291 https://www.verizonwireless.com/aboutus/commitment/safety-security/device-unlocking-policy.html 292 https://www.verizonwireless.com/aboutus/commitment/safety-security/device-unlocking-policy.html 293 http://arstechnica.com/gadgets/2014/08/t-mobile-becomes-first-american-carrier-to-release-phone-unlocking-app/ 294 Federal-State Joint Board on Universal Service, TracFone Wireless, Inc. Petition for Designation as an Eligible Telecommunications Carrier in the State of New York et al., CC Docket No. 96-45, Order, 23 FCC Rcd 6206, 6212 n.37 (2008) (TracFone ETC Designation Order) (“TracFone committed that it will comply with the Consumer Code for Wireless Service of the CTIA.”); see also 47 C.F.R. § 54.202 (allowing ETCs to demonstrate commitment to consumer protection by committing to comply with CTIA’s Consumer Code for Wireless Services). 295 http://www.tracfone.com/includes/content/popup/unlocking_policy.jsp?a=1296504971589 65 Federal Communications Commission DA 14-1862 Cellular (850 MHz band) and PCS (1.9 GHz band) spectrum. Similarly, service providers are deploying LTE on different spectrum bands. For example, AT&T launched LTE using Lower 700 MHz B and C block spectrum while Verizon Wireless launched LTE using the Upper 700 MHz C block spectrum.296 Other providers holding primarily Lower 700 MHz Band A Block licenses filed a petition for rulemaking asking the Commission to require that all mobile units for the 700 MHz band be capable of operating over all frequencies in the band.297 In September, 2013, a number of the principal wireless providers licensed in this band, along with the Competitive Carriers Association, developed a voluntary industry solution to resolve the lack of interoperability in the Lower 700 MHz band while allowing flexibility in responding to evolving consumer needs and technological developments.298 In October, 2013, the Commission adopted a Report and Order and Order of Proposed Modification to effectuate the industry solution and thereby provide for interoperable LTE service in the Lower 700 MHz band.299 Since October, 2013, the Commission has adopted specific interoperability requirements for the AWS-3 band,300 as well as for the 600 MHz Band, which is to be auctioned as part of the broadcast incentive auction.301 V. PRICING LEVELS AND TRENDS 126. Among the most significant developments in mobile pricing during the period under review have been changes in the two major pricing models traditionally used in the United States. Most mobile telephone subscribers are billed monthly for their mobile wireless service after the service has been provided (“postpaid” service). In contrast, other mobile subscribers, including those lacking the necessary credit history, are required to pay for their service in advance (“prepaid service”). Historically, the terms “postpaid” and “prepaid” service were largely synonymous with “contract” and “no-contract” pricing plans. This is because the handset subsidy model traditionally used in postpaid service requires customers to sign a contract for a specified period (typically two years) in return for receiving a significant upfront discount on the price of a handset, with service providers recovering the balance of the handset cost over the course of the contract through the higher monthly fees they charge for mobile wireless services.302 As noted earlier, there has recently been a significant increase in new service plans that employ a different, “no-contract” postpaid model, due primarily to the offering of installment payment plans and a separation of service and equipment fees. Thus, some of the distinctions between the two service offerings have diminished. There are now more postpaid plans with no handset subsidies or service contracts than were previously available. In addition, similar pricing plan options are increasingly available to customers of both models. 127. While the lines have been blurred somewhat, there remain certain features, such as access to 296 See Section IV.A, supra 297 700 MHz Block A Good Faith Purchaser Alliance Petition for Rulemaking Regarding the Need for 700 MHz Mobile Equipment to be Capable of Operating on All Paired Commercial 700 MHz Frequency Blocks, filed Sept. 29, 2009 (700 MHz Equipment Petition), at iii, 12. 298 [cite early September filings of parties in the proceeding.] 299 In the Matter of Promoting Interoperability in the 700 MHz Commercial Spectrum, Requests for Waiver of Lower 700 MHz Band Interim Construction Benchmark Deadlines, WT Docket Nos. 12-69, 12-332, Report and Order and Order of Proposed Modification, 28 FCC Rcd 15122 (2013). As a final step implementing the industry solution, the Commission issued an Order in January, 2014, modifying AT&T’s Lower 700 MHz B and C Block licenses consistent with the 700 MHz Interoperability Report and Order and Order of Proposed Modification. See In the Matter of Promoting Interoperability in the 700 MHz Commercial Spectrum, WT Docket No. 12-69, Order of Modification, 29 FCC Rcd 281 (2014). 300 See AWS-3 Report and Order at , ¶¶ 225-231. 301 See Incentive Auctions Report and Order at ¶¶ 731-737. 302 Alexia Gonzalez Fanfalone and Agustin Diaz-Pines, Mobile Handset Acquisition Models, Organization for Economic Cooperation and Development, July 5, 2013, at 1-7. 66 Federal Communications Commission DA 14-1862 roaming coverage, means of enforcing data allowance limits, and, importantly, the requirement of a credit check, that continue to differentiate the two models. The following discussion of developments in mobile service pricing plans addresses these issues, and is divided into two sections. The first section covers developments in postpaid plans. The second section covers developments in prepaid plans, including traditional prepaid service plans and higher-end prepaid service plans that include data. The discussion focuses on recent changes made by providers during the period covered by this Report. It does not present a comprehensive comparison of pricing plans and pricing data.303 A. Postpaid Service 128. In the period since the Sixteenth Report, the most significant development in postpaid service has been the rise of the equipment installment plan (EIP) as an alternative to the traditional handset subsidy model. As discussed below, the growth of EIPs was accompanied by the introduction of discounted no-contract service plans for customers who forego handset subsidies, and this reduction in the price of non-subsidized postpaid service plans has been accompanied by other types of pricing changes. The following section focuses on the rise of EIPs and its effects, both direct and indirect, on the pricing of postpaid service. Other key developments in postpaid service, including the continuing evolution of data pricing models and new services that exempt certain types of data usage from counting against a mobile subscriber’s monthly data allowance, are also examined at the end of the section. 1. Equipment Installment Plans 129. During the period covered by the Sixteenth Report, there has been an increase in the availability of different types of Equipment Install Plans (EIPs) for postpaid services. For instance, early in this period, T- Mobile lowered monthly service fees on its highest tiered data plans for customers who brought their own handset, paid full price up-front, or signed a contract to pay for a new handset in monthly installments. While other nationwide providers were already offering a month-to-month no-contract option to customers who brought their own device or paid the full price of a new device upfront, they did not offer any reduction in monthly service fees or an EIP financing option. 130. In March 2013, T-Mobile ceased offering traditional contract plans with handset subsidies to new customers and shifted exclusively to a no-contract model with discounted monthly service fees for customers who bring their own handset, pay the full price upfront, or sign a contract to pay for a handset in monthly installments. Since then, the other three nationwide providers and some regional providers have all introduced their own version of an installment payment option and a discounted, no-contract postpaid service plan.304 To date, however, T-Mobile remains the only nationwide provider that has completely eliminated handset subsidies and shifted exclusively to the no-contract version of the postpaid model. 131. While the other three nationwide providers continue to offer traditional contract plans with handset subsidies alongside the no-contract EIP option, all four nationwide providers have structured customer incentives to encourage the adoption of EIPs.305 Beginning with T-Mobile, all four nationwide providers and 303 While mobile pricing plans have become increasingly national in scope, the geographic coverage of any particular pricing plan or offer may vary across regional markets for a number of reasons. In some cases, service providers may not offer certain broadband data plans in geographic markets where they have not yet upgraded their networks. In other cases, service providers conduct pilot tests of new pricing plans (or changes in existing pricing plans) in selected regional markets before offering them across the rest of their network footprint. We have not attempted to make a systematic determination of the geographic availability of the pricing plans and rates covered in this section of the Report on a market-by-market basis. Therefore, we cannot state with certainty that all the particular pricing plans and rates covered by the Report are available in any given geographic market. See, e.g., AT&T-Leap Order, 29 FCC Rcd at 2749 ¶ 30,31; SoftBank-Sprint Order, 28 FCC Rcd at 9657 ¶ 38 304 See Appendix Tables V.A.i and V.A.ii for details 305 David W. Barden et al., The Rise of Installment Payment Plans and Implications for Wireless Carriers, Bank of America Merrill Lynch, Equity Research, Feb. 7, 2014, at 2, 4-5. 67 Federal Communications Commission DA 14-1862 some regional providers have introduced early handset upgrade plans that link the option to upgrade more frequently to EIPs. Although the details of early upgrade plans vary across providers, all such plans allow customers to upgrade earlier and more frequently than allowed under traditional subsidized plans, but require customers who take advantage of this early upgrade option to pay for their new handsets with an EIP. 132. A second way service providers have encouraged EIP adoption is by offering lower monthly service fees to customers who pay for handsets in monthly installments. While AT&T and Verizon initially limited their efforts to incentivize EIP adoption to their early handset upgrade plans, they began offering such service fee discounts within a year if introducing EIPs,followed by a subsequent round of greater discounts. As noted below, when Sprint introduced new shared data plans in August 2014, it offered discounts for EIP customers that are similar to those offered by AT&T and Verizon. In addition, a subsequent price reduction on the unsubsidized version of Sprint’s unlimited data plan for individuals effectively increased the discount available to EIP customers for this particular plan.306 133. From the outset, most of the nationwide providers – T-Mobile, AT&T, and Sprint – consistently offered the same discounts available for EIP customers to other customers who pay for their own handsets in other ways, including customers who pay the full price of a new handset upfront, new customers who bring their own devices (“BYOD”), and existing customers who have already paid for their current handset under a traditional two-year contract that has expired. In contrast, when Verizon first started offering EIP discounts, the company initially limited eligibility to customers on its “Edge” EIP/upgrade plan: it only subsequently extended these same discounts to existing customers on month-to-month contracts and new customers who bring their own devices. In addition, while most providers have offered discounts exclusively to customers who pay for their own devices in one of these ways, in February 2014 AT&T extended the same discounts it offers on non-subsidized no-contract plans to its existing subsidized contract customers – in other words, customers who had not yet shifted to EIPs -- until their next handset upgrade. 134. As noted above, the reduction in the price of non-subsidized no-contact plans was accompanied by other pricing changes that effectively reduced the price of traditional subsidized contract plans along with the new no-contract plans. These other types of pricing changes are examined in greater detail below. 2. Changes in Monthly Pricing of Postpaid Plans 135. In addition to offering discounts on service fees to EIP customers and others who forego handset subsidies, service providers made a number changes to the monthly pricing of postpaid service plans that effectively cut the price of postpaid service for customers whose data use had, previously, either put them at risk of overages or would have necessitated subscribing to the next highest tier. In some cases, both AT&T and Verizon cut the monthly service fees on selected data tiers outright. More commonly, some providers, including Verizon, T-Mobile and U.S. Cellular, effectively reduced the price of postpaid service plans by increasing monthly data allowances on usage-based data plans without increasing monthly service fees. In addition, some providers increased the value of existing plans by adding new features, including international text messaging and cloud storage allowances. 306 It’s a New Day for Unlimited Data, Press Release, Sprint, Aug. 21, 2014; Mike Dano, Sprint Drops Unlimited Everything Plan to $60, Undercuts T-Mobile by $20, FIERCEWIRELESS, Aug. 21, 2014. 68 Federal Communications Commission DA 14-1862 T-Mo Simple Choice "Un-carrier 1.0" 2 Yr. contract discontinued; Launch no contract plans; EIP Sprint Unlimited/ My Way/ My All-in Launch of fixed price unlimited talk/text/data plans AT&T Mobile Share Traditional plans discontinued. Customers much sign up for shared plans VZW Share Everything Launch of less expensive entry level plans AT&T Mobile Share Monthly service fee based on usage and number of smart phone lines VZW More Everything Data allowances increased and fees reduced VZW Loyalty Plans Cheaper price plans to retain customers T-Mo Simple Choice Data allowance increased, Unlimited data price increased AT&T Mobile Share Reduced monthly access fee and data price U.S. Cellular Data allowance increased T-Mo Simple Starter Lauch entry level LTE plan T-Mo All Plans Overage charges abolished VZW More Everything Shared business plans launched Chart V-1 Major Changes to Pricing of Postpaid Service Plans 2013- December 2014 T-Mo "Un-carrier 3.0" International Roaming charges dropped AT&T Mobile Share/ Value Reduced data price, unlimited international Sprint Family Share Pack $100 for 28 GB of shared data for upto 4 T-Mo Simple Choice Unlimited 4G LTE Family Plan $100 for 2 lines 69 Federal Communications Commission DA 14-1862 136. During this period, Sprint limited its response to providing discounts for EIP customers and other non-subsidized subscribers through its Framily plans. In August 2014, Sprint replaced its Framily plans with a new shared data offering, called Sprint Family Share Pack.307 Data pricing under the new plans is lower than Sprint’s previous data pricing, and is either in line with or substantially lower than that of Verizon’s and AT&T’s shared data offerings depending on the usage level and whether the plan is subsidized or unsubsidized.308 The key difference between Sprint’s shared data offering and those of Verizon and AT&T lies in the size of the monthly data allowances -- most of Sprint’s shared data tiers offer at least double the data allowance of similarly-priced shared data tiers from AT&T and Verizon. 137. Sprint followed up shortly thereafter by reducing the price of its unlimited data plan for individuals to $60 per month, down from $80 per month for the subsidized version of its unlimited data plan and $75 per month for the unsubsidized version.309 The plan is available to both new and existing Sprint customers, but customers must purchase their device through Sprint’s EIP plan (Easy Pay), pay full retail price, or bring their own compatible device to qualify for the plan. 138. For Verizon and AT&T, price cuts on postpaid service plans generally benefited both subscribers on traditional subsidized contract plans and those on unsubsidized no-contract plans, inasmuch as both providers set the discounts on no-contract plans that are similar to the prices of the corresponding traditional contract offerings.310 Similarly, when Sprint introduced its new shared data plans, it effectively reduced prices for both subsidized and unsubsidized plans. In contrast, since T-Mobile no longer offers traditional contract plans with handset subsidies, its postpaid subscribers could get a price reduction only by signing up for one of the new non-subsidized no-contract plans. Just recently in early December, T-Mobile introduced an unlimited 4G LTE family plan for $100 for 2 lines.311 139. While price cuts were the predominant form of changes to the monthly pricing of postpaid service plans, there were exceptions. In March 2014, for example, T-Mobile increased the price of its unlimited data plan by $10 per month, though it also doubled the tethered data allowance on this same plan to 5GB at the same time. 307 It’s a New Day for Data for American Consumers, Press Release, Sprint, Aug. 19, 2014. 308 Jonathan Chaplin et al., Sprint Focuses Value Proposition on More Data Rather Than Lower Pricing, New Street Research, Equity Research, Aug. 19, 2014, at 2-4; Phil Goldstein, Analysts: Sprint’s New Shared Data Plans Not Disruptive Enough to Change its Fortunes, FIERCEWIRELESS, Aug. 19, 2014. Sprint charges the same monthly line access fee per smartphone for traditional subsidized contract plans as AT&T and Verizon, and the discounts on this line access charge offered for unsubsidized no-contract plans are similar to those offered by AT&T and Verizon. 309 It’s a New Day for Unlimited Data, Press Release, Sprint, Aug. 21, 2014; Mike Dano, Sprint Drops Unlimited Everything Plan to $60, Undercuts T-Mobile by $20, FIERCEWIRELESS, Aug. 21, 2014. The plan also includes unlimited voice and texting. 310 Both Verizon and AT&T discount their shared data plans for EIP customers by reducing the monthly access fee for adding a smartphone line, rather than the monthly fee for each tier’s shared data allowance. Accordingly, when they reduce the monthly fee for the shared data allowance of a specific data tier or increase the shared data allowance of a specific data tier at existing price levels, the pricing change benefits both subsidized and non-subsidized customers. 311 http://newsroom.t-mobile.com/news/unlimited-family-plan.htm (accessed December 11, 2014) 70 Federal Communications Commission DA 14-1862 T-Mo Jump Upgrade every 6mo. for $10/ mo. AT&T Next Upgrade every 12 mo. No downpayment VZW Edge Upgrade every 6 mo. if 50% device cost paid off Sprint One Up Upgrade after 12 mo. No downpayment Sprint Family Plans/ Easy Pay Upgrade after 12 mo. on 24 mo. EIP. Early upgrade with unlimited data plan VZW Edge Upgrade after 30 days if 50% device cost paid off T-Mo Jump Upgrade anytime as often as they want if 50% device cost paid off. VZW Edge Make upgrade conditions stricter Jul-13 Aug-13 Sep-13 Oct-13 Nov-13 Dec-13 Jan-14 Feb-14 Mar-14 Apr-14 May-14 Jun-14 Jul-14 Chart V.A.2 Installment Payment Plans with Early Upgrade Options 2013- August 2014 Sprint One Up $15/Mo per line for Unlimited plans AT&T Mobile Share Value/ Next $15/mo per line Sprint Framily Plan/ Easy Pay larger discount as more people join plan, max $10/mo per line AT&T Mobile Share Value/ Next Additional $10/mo per line for 10 GB data or more VZW More Everything/ Edge $10 - $20 / mo. per line for 8 -10 GB data or more C Spire Unlimited Everything $35/mo existing unlimited voice & data plans VZW More Everything Discounts extended to existing customers on mo. to mo. contracts C Spire Shared Data $15/ mo. per line for 8 GB or less; $25/ mo. per line for 10 GB or more Sep-13 Oct-13 Nov-13 Dec-13 Jan-14 Feb-14 Mar-14 Apr-14 May-14 Jun-14 Chart V.A.3 Discounts on Monthly Service Fees for EIP Customers 2013- August 2014 VZW More Everything $25/mo. per line for 10 71 Federal Communications Commission DA 14-1862 3. Impact of EIPs on Consumer Costs 140. As discussed above, service providers now offer lower monthly service fees to customers who opt to pay for smartphones in monthly installments. Discounts typically range from $10 to $25 per month, relative to traditional contract plans with handset subsidies, depending on what data tier customers choose. As various industry analysts note, the overall cost of postpaid service to consumers under these plans depends, on a consideration of the costs of both the handset and the service plan. As one report points out, 312 installment payment plans have also reduced handset subsidies, and the subsidy reduction offsets the reduction in monthly service fees over the life of the customer. Different analysts have attempted to determine the net effect of installment plans on the total cost facing consumers both for service and the handset. The results vary, due in part to differences in method, underlying assumptions and the way analysts interpret industry developments. 141. Macquarie Research estimated that consumers who choose the no-contract installment payment option pay $74 more per year on average than consumers on traditional contract plans with embedded handset subsidies.313 Under Macquarie’s assumptions, the reduction in the handset subsidy raises the total cost to the consumer by more than the reduction in the price of the service plan lowers the total cost.314 On the other hand, because customers on installment payment plans are no longer under contract, they are free to switch providers whenever they want.315 142. Another analysis by Bank of America Merrill Lynch suggests that the net effect of installment payment plans on the total cost facing the consumer varies depending on the service provider and the type of service plan.316 According to this report, the first $10-15 of monthly service price reductions for EIP subscribers is essentially the shift from service to equipment revenue, but discounts over and above this threshold represent real price reductions. The report concluded that an installment payment plan is a cheaper option than receiving a subsidized device over a 24 month period for Verizon’s and AT&T’s plans with ten or more gigabytes of shared data, where the service price discounts offered for taking the installment payment option are higher than the discounts offered with lower data tiers.317 143. The conclusion of an analysis by New Street Research is that, installment plans represent a price cut despite the reduction in the handset subsidy.318 New Street Research has estimated that the monthly price of 312 Jonathan Chaplin et al., 1Q14 Wireless Trends Review: Competition Takes Toll; Pressures Likely to Intensify in Coming Quarters, New Street Research, Equity Research, May 22, 2014, at 20. 313 Jay Yarow, U.S. Wireless Carriers are About to Change How People Buy Smartphones, BUSINESS INSIDER, May 15, 2014. 314 Macquarie assumed that the discount on an LTE data plan for the installment payment option is $10 per month, which is at the low end of the range of currently available discounts. Macquarie also estimated that customers on installment payment plans pay $28 more per month on average and that their total payments are $1,434 more on average over an eight-year period. 315 Jay Yarow, U.S. Wireless Carriers are About to Change How People Buy Smartphones, BUSINESS INSIDER, May 15, 2014. 316 David W. Barden et al., The Numbers, the Comparable Numbers, and What’s Next …, Bank of America Merrill Lynch, Equity Research, May 27, 2014, at 7-8. 317 David W. Barden et al., The Numbers, the Comparable Numbers, and What’s Next …, Bank of America Merrill Lynch, Equity Research, May 27, 2014, at 7-8. Both Verizon and AT&T offer a discount of $25 per line per month to customers who take the installment payment option with data tiers of 10GB or more; for data tiers of less than 10GB, AT&T offers a $15 per line monthly discount, and Verizon offers a $10 per line monthly discount. The analysis assumes that four lines share the data tiers with at least 10GB. 318 Jonathan Chaplin et al., 1Q14 Wireless Trends Review: Competition Takes Toll; Pressures Likely to Intensify in Coming Quarters, New Street Research, Equity Research, May 22, 2014, at 13. 72 Federal Communications Commission DA 14-1862 representative service plans in May 2014 decreased by an average of 20 percent over the previous year.319 According to this analysis. most if not all of the cost savings due to lower handset subsidies have been offset by ETF buyouts and increased tablet subsidies, as well as a rise in sales and marketing expenditure, while the price of service plans is unlikely to rise.320 4. Promotions and Incentives 144. Historically, handset subsidies have been an important marketing tool for customer acquisition and retention in the U.S. mobile wireless market.321 With the shift away from handset subsidies, service providers have increasingly offered a wide variety of other types of promotions and incentives in an effort both to attract and retain customers and encourage increased usage of mobile wireless services, especially data.322 The discussion in this section focuses on two of the most widely used types of offers -- Early Termination Fee (ETF) buyouts and tablet promotions and subsidies. 145. The purpose of ETF buyouts is to encourage customers to switch from rivals by reducing switching costs. To this end, ETF buyouts typically include a cash payment or credit to reimburse ETFs (or pay off the remaining balance on an EIP) plus a separate device credit for trading in the customer’s current handset. . The first ETF buyout offer came from AT&T which for one month offered up to $450 in credit to customers who switched from T-Mobile to AT&T and traded in their current devices. The offer was limited to customers who pay for their handset via EIP, full price upfront, or BYOD. Shortly thereafter, T-Mobile offered to reimburse up to $350 in ETFs and a $300 device credit for phone trade-in to customers who switched from any of the other three nationwide providers to T-Mobile. In April 2014, Sprint offered to reimburse up to $650 in switching costs for customers who switch their number to Sprint from another postpaid provider and sign up for Sprint’s non- subsidized, no-contract “Framily” plan. Regional providers such as C Spire and U.S. Cellular made similar buyout offers. Sprint offered yet another ETF buyout in conjunction with the launch of its new shared data plans in August 2014. While AT&T’s buyout offer lasted one month and Sprint has stated that its offer is available for an undefined, limited time basis, T-Mobile has said that it has no plans to end its ETF buyout.. 146. During this period, many providers also offered various incentive plans for customers purchasing tablets that use cellular networks in addition to Wi-Fi access. For example, T-Mobile offered lifetime access to 200MB of free LTE data per month to all customers who use tablets on their network, and Verizon offered 1GB of extra data per month free of additional charge to customers who activate a tablet on its shared data plans at or above the 1GB data tier. Providers have also tried to encourage increased tablet data usage on their networks by subsidizing tablets. Sprint offered a free Samsung tablet to customers who join its no-contract “Framily” plan if they also sign up for a qualifying data plan. Sprint’s free tablet offer began as a short-term promotion, but was extended multiple times in the first quarter of 2014. Similarly, Verizon retail stores have been offering customers a $100 discount off the price of tablets with a two-year contract, and some Verizon stores have offered selected tablets for either $50 or free of charge with a two-year contract.323 T-Mobile temporarily offered LTE-enabled 319 Jonathan Chaplin et al., 1Q14 Wireless Trends Review: Competition Takes Toll; Pressures Likely to Intensify in Coming Quarters, New Street Research, Equity Research, May 22, 2014, at 13. The estimate is for representative 3-line plans. For prior plans, the estimate assumes unlimited voice and text, with approximately 2GB of data per subscriber for AT&T and Verizon, 2GB for T-Mobile, and unlimited data for Sprint. For current plans, the estimate assumes unlimited voice and text, with 2GB per subscriber for AT&T and Verizon, and 3GB for Sprint and T-Mobile. 320 Id., at 20-22. 321 Jonathan Chaplin et al., 1Q14 Wireless Trends Review: Competition Takes Toll; Pressures Likely to Intensify in Coming Quarters, New Street Research, Equity Research, May 22, 2014, at 20. See also, Alexia Gonzalez Fanfalone and Agustin Diaz-Pines, Mobile Handset Acquisition Models, Organization for Economic Cooperation and Development, July 5, 2013, at 1, 8. 322 see Appendix Table V.A.iii for details 323 Mike Dano, Verizon’s Comprehensive Tablet Strategy is Leading the Way, FierceWireless, July 23, 2014. 73 Federal Communications Commission DA 14-1862 tablets for the same price as Wi-Fi-only models for any postpaid activation of a mobile internet data plan of 1GB or more In early December 2014, Sprint came out with a promotional offer to cut consumers’ rate plans in half if they switch from Verizon Wireless or AT&T to Sprint. This offer would be good for as long as they stay a Sprint customer.324 324 http://newsroom.sprint.com/blogs/devices-apps-and-services/sprint-ceo-marcelo-claure-bold-move-cuts-wireless-bills-in- half.htm 74 Federal Communications Commission DA 14-1862 Sprint Unlimited/ My Way/ My All-in Unlimited data to new customers AT&T New tablet data plan with free 1 GB data T-Mo "Un-carrier 4.0" Lifetime free 250MB LTE data per mo. with tablet plan Sprint Framily Plan Free Galaxy Tab3 with plan signup AT&T Upto $450 to customers who switch from T-Mobile T-Mo Upto $650 paid to customers who switch from AT&T AT&T Mobile Share $100 credit for new lines VZW $35 activation fee waived C Spire $200 service credit to customers who switch VZW More Everything Extra data per month for tablets with Sprint Framily Plan Upto $650 paid to customers who switch U.S. Cellular Share Connect Upto $350 paid to customers who switch T-Mo LTE tablets same price as WiFi only; $10 discount on data plans T-Mo "Un-carrier 5.0" Customers can use iPhone 5s with unlimited service at no cost for 7 days, Chart V-4 Promotions and Incentives to Attract and Retain Customers 2013- December 2014 Sprint Cut Your Bill in Half Cuts rate plan in half when you switch from VZW or AT&T 75 Federal Communications Commission DA 14-1862 5. Evolution of Data Pricing Models 147. In addition to the shift to EIPs and associated pricing changes, the data pricing models that the nationwide providers use to differentiate their service offerings continued to evolve during the period covered by this Report. The overall trend was toward somewhat greater convergence of data pricing models than before.325 148. As noted in the Sixteenth Report, three distinct data pricing models had emerged in 2011 – the tiered, usage-based data pricing model with overage charges adopted by AT&T and Verizon; the tiered, usage- based model with speed reductions adopted by T-Mobile; and the unlimited data pricing model used by Sprint. There were two major changes to these pricing models in the second half of 2012. First, both AT&T and Verizon launched shared data plans that allow customers to pool their monthly data allowance across multiple smartphones, tablets, and other devices, rather than having to purchase a separate data plan for each device.326 Second, T-Mobile reintroduced an unlimited data plan alongside its usage-based data tiers. Sprint remained the only nationwide provider that continued to offer exclusively an unlimited data plan. 149. Since the period covered by the Sixteenth Report, the most significant change for Sprint was its introduction of usage-based data plans, although its unlimited data pricing model remains a significant part of its pricing strategy. Sprint first launched an entry-level, usage-based data tier with overage charges alongside its unlimited data plans in the second half of 2013, and subsequently included two usage-based data tiers with overage charges in the Framily plans for EIP customers that it introduced at the beginning of 2014. In July 2013 it launched a promotion guaranteeing unlimited data for the life of the line to new and existing customers who sign up for one of its unlimited data plan options. In August 2014, Sprint replaced it usage-based Framily plans with the new Family Share Pack shared data plans discussed above, and cut the price of its unlimited data plan for individuals.327 150. During this time, AT&T also modified its data pricing in ways that made it more similar to Verizon’s pricing. When shared data plans were first introduced in 2012, Verizon ceased offering its existing individual tiered data plans while AT&T continued to offer both individual and shared data plans. In addition, whereas Verizon charged a standard monthly fee ($40 per month) to add a smartphone line for all data tiers, AT&T instituted a sliding scale of fees ($30-50 per month) depending on the data tier. In the second half of 2013, AT&T discontinued its legacy tiered data plans, and subsequently established a standard monthly fee of $40 per month to add a smartphone line in place of the original sliding scale of fees. 151. T-Mobile continues to differentiate its general data pricing model based on the use of speed reductions, rather than overage charges, when data users reach the limit of their monthly data allowance. In May 2014, T-Mobile abolished remaining overage charges on all plans for domestic calls, text messages and data usage. T-Mobile also launched a new entry-level LTE data plan in April 2014, where service is suspended entirely once the customer reaches the 500MB monthly LTE data allowance, with the customers having the option to purchase a temporary data pass for an extra charge. 328 . 325 See Appendix Table V.A.iv for details. In addition, Appendix Table V.A.v presents selected basic and smartphone plans for the four national and selected regional providers that are representative of the pricing changes in the marketplace 326 At the time Verizon started offering its shared data plan, it also ceased offering unlimited data plans to new customers. 327 In conjunction with the launch of this new unlimited plan, Sprint discontinued its promotional offer guaranteeing unlimited data for the life of the line to new customers who sign up for certain unlimited data plans. Sprint’s new unlimited data plan does not come with this guarantee, and existing customers who already have that benefit on their account must stay on the rate plan that includes the lifetime guarantee in order to retain the benefit. Phil Goldstein, Sprint Kills Unlimited Data for Life Guarantee for New Customers, FierceWireless, Aug. 21, 2014. 328 Appendix Table V.A.v presents selected basic and smartphone plans for the four national and selected regional providers that are representative of the pricing changes in the marketplace 76 Federal Communications Commission DA 14-1862 6. Other Developments 152. During this period, certain mobile providers have implemented plans and services that exclude certain types of data usage from counting against a customer’s monthly data allowance. This effectively lowers the price of data for any consumers who use the affected services and who exceed or may exceed their data allowance based on the use of those services. In January 2014, AT&T announced data plans, called “Sponsored Data,” that will allow content providers to pay for mobile subscribers’ data usage.329 In addition, T-Mobile’s recently announced “Music Freedom” unlimited free streaming music service allows customers unlimited access to six music streaming services without using up any of the data allowance on their 4G plan.330 153. Another recent development is the launch of more attractive international roaming plans. T- Mobile is currently the only provider that does not charge international roaming fees for certain services in more than 100 countries through its Simple Global plan. Individual and business customers on the Simple Choice plan receive unlimited 2G (Edge) data and texting globally.331 As of February 2014, AT&T has adopted 4G LTE data roaming in 15 countries, and is still the only US provider to have 4G LTE speeds abroad.332 B. Prepaid Service 154. The four nationwide providers offer their own prepaid service, in addition to contracting with MVNOs who then offer service on those nationwide networks. As identified in previous Reports, analysts believe that it is appropriate to split the prepaid market into a low-end segment and a high-end segment.333 The low-end segment involves traditional pay-as-you-go service, while the high-end segment encompasses unlimited and tiered usage-based plans.334 All national providers serve both the low-end segment and the high- end segment.335 Of the major players in the prepaid segment, AT&T offers prepaid and pay-as-you-go plans under its GoPhone brand,336 and also offers higher-end prepaid plans under the Cricket brand name.337 T- Mobile offers a selection of prepaid plans, including both pay-as-you-go plans and unlimited higher end plans under the T-Mobile, MetroPCS, and GoSmart brand names. Depending on the brand name and plan structure, these plans include varying allotments of monthly 4G LTE data, along with unlimited data at slower speeds. T- Mobile offers the same monthly plans for its monthly prepaid customers as it does to its postpaid customers, although, as described below, certain other differences remain between the offerings for the two sets of customers. Sprint offers several prepaid brands on its network, all with both smartphone and feature phone plans, and each with a different target audience. Boost Mobile338 serves subscribers who are voice and text Mike Dano, Hershey’s, Cut the Rope Among Advertisers Using AT&T’s Sponsored Data Service, FierceWireless, June 25, 2014; Phil Goldstein, AT&T Sponsored Data Partner Syntonic Wireless to Launch Toll-Free Content Store, FierceWireless,, July 7, 2014. 330 Sue Marek, T-Mobile Entices Consumers to “Cheat on their Wireless Carrier’ With7-Day Free iPhone Trial, FierceWireless, June 19, 2014. 331 http://www.fiercewireless.com/story/t-mobile-kills-international-data-roaming-fees/2013-10-09 (visited June 26, 2014) 332 In December, AT&T launched LTE data roaming in Canada and the United Kingdom. Now, AT&T has made those speeds available to Spain, France, Japan, South Korea, Australia, Hong Kong, Switzerland, Malaysia, Russia, Singapore, Guam, Hong Kong and Antigua & Barbuda. AT&T plans to continue expanding its LTE footprint. The provider has agreements allowing LTE roaming in 200 countries; See http://www.fiercewireless.com/press-releases/att-launches-lte- roaming-more-countries-remains-only-us-carrier-offer-inter-0 (visited July 8, 2014). 333 Need cite 334 Sixteenth Report, 28 FCC Rcd at ¶ 160-161. 335 See Appendix V.B for detailed information of selected pre-paid plans. 336 http://www.att.com/shop/wireless/gophone.html 337 https://www.cricketwireless.com/cell-phone-plans 338 http://www.boostmobile.com/ 77 Federal Communications Commission DA 14-1862 messaging-centric. It also provides a plan with a price that falls as monthly payments are repeatedly made on time. Virgin Mobile339 serves subscribers who are device and data-oriented, and also provides service for the Lifeline program under the Assurance Wireless brand.340 Verizon Wireless offers basic pay-as-you-go341 and daily prepaid plans,342 only for use with feature phones, as well as monthly prepaid plans.343 155. The Sixteenth Report noted several key trends in prepaid service pricing during that period.344 First, nationwide service providers increased their presence in the prepaid market segment by launching their own prepaid brands in competition with resellers and multi-metro prepaid service providers such as Leap and MetroPCS. Second, prepaid service providers expanded their mobile broadband data service offerings for smartphones and other connected devices. Third, prepaid mobile broadband pricing plans shifted from unlimited data pricing to tiered, usage-based data pricing coupled with speed reductions after usage exceeds a monthly allowance for high-speed data. 156. The structure of the prepaid market continued to evolve during the period covered by the this Report. The prepaid service segment has witnessed robust growth , and a substantial proportion of mobile customers in the United States now purchase their service on a prepaid basis.345 Consumers of these prepaid plans may be less frequent users, lack the credit history necessary for postpaid service, or may prefer the flexibility and value of prepaid plans. Higher end prepaid plans are generally purchased on a per-month or per- day basis, while lower-end prepaid plans are generally those in which the customer incrementally reloads their account with data, voice, and text. 157. While one continuing trend in the prepaid service segment has been the offering of some prepaid pricing plans very similar to postpaid plans, there are significant differences in non-price plan features of prepaid plans – although the limitations vary by service provider and by plan. The main distinctions, apart from the issue of a credit check for customers, tend to be more limited data speeds, less extensive geographic coverage, more limited usage allowances, and more limited handset options. For instance, while some plans might include 4G (and some limited LTE) data, many prepaid plans restrict users to 2G/3G speeds. Certain providers restrict their prepaid customers from accessing roaming networks. Providers may also prioritize their postpaid traffic over their prepaid traffic, leading to possible quality degradation for the prepaid service. For example, T-Mobile prioritizes T-Mobile-branded traffic over that of its MetroPCS and GoSmart Mobile prepaid brands on its HSPA and LTE network.346 339 http://www.virginmobileusa.com/ 340 http://www.sec.gov/Archives/edgar/data/101830/000010183014000012/sprintcorp201310-k.htm 341 http://www.verizonwireless.com/pre-paid/pay-as-you-go/ 342 http://www.verizonwireless.com/wcms/consumer/shop/pre-paid.html?t=2 343 http://www.verizonwireless.com/wcms/consumer/shop/pre-paid.html?t=2 344 Sixteenth Report, 28 FCC Rcd at ¶161-173. 345 As of June 2014, prepaid customers accounted for approximately 34 percent of mobile customers in the United States, based on ComScore MobilLens 3 month average survey data. 346 Mike Dano, AT&T Caps Cricket’s Peak Download Speeds to 8 Mbps on LTE, 4 Mbps on HSPA+, FIERCEWIRELESS, May 20, 2014. 78 Federal Communications Commission DA 14-1862 AT&T Aio launch T-Mobile MetroPCS brand launched nationwide VZW Increases data allowance AT&T Increases Aio coverage area and adds LTE T-Mobile Adds entry level plan MetroPCS plan at $40, $100 for 4 lines AT&T Aio launched nationwide T-Mobile Brightspot launched as new pre-paid offering Sprint Price cuts in Boost LTE plans from $55 to $35 VZW Allset plans reduce price and data allowance Sprint New brand and plan launched at $45 and $60 Sprint Speeds reduced for heavy users AT&T Aio group discount plans launched Sprint Price cuts of unlimited plans for Boost AT&T New Cricket, New plans starting at $40, $100 for 4 lines, Aio shut down T-Mobile Price cut on high-end and other data plans May-13 Jun-13 Jul-13 Aug-13 Sep-13 Oct-13 Nov-13 Dec-13 Jan-14 Feb-14 Mar-14 Apr-14 May-14 Jun-14 Jul-14 Chart V.A.5 Major Changes to Pricing of Prepaid Service Plans 2013- July 2014 79 Federal Communications Commission DA 14-1862 AT&T Aio launch T-Mobile MetroPCS brand launched nationwide VZW Increases data allowance AT&T Increases Aio coverage area and adds LTE T-Mobile Adds entry level plan MetroPCS plan at $40 AT&T Aio launched nationwide T-Mobile Brightspot launched as new pre-paid offering Sprint Price cuts in Boost LTE plans from $55 to $35 VZW Allset plans reduce price and data allowance Sprint New brand and plan launched at $45 and $60 AT&T Aio group discount plans Sprint Price cuts of unlimited plans for Boost AT&T Aio shut down- Launched as New Cricket, New plans starting at $40 AT&T Support of Govt. Lifeline program phased out Chart V-5 Major Changes to Pricing of Pre-paid Service Plans 2013- August 2014 T-Mo Price cut on high-end and 80 Federal Communications Commission DA 14-1862 158. In addition, during the period under review, the four nationwide providers took significant steps in their offering of prepaid services, including acquisition of the two largest facilities-based prepaid service providers. In early 2013, T-Mobile acquired MetroPCS, and soon after, in May 2013, AT&T launched its nationwide prepaid brand Aio, in selected markets.347 In July 2013, T-Mobile launched the MetroPCS brand in 15 new metro areas,348 increasing the MetroPCS footprint by 50 million POPs.349 In early 2014, AT&T acquired Leap (marketed as Cricket Wireless)350 and launched it as the New Cricket in May 2014 while simultaneously shutting down its existing Aio Wireless prepaid brand.351 Key developments in the prepaid market are discussed below and shown in Chart V.A.5.352 1. Prepaid Plan Choices 159. Since early 2013, the nationwide providers have continued to offer both value-conscious low-end prepaid plans, as well as higher-end prepaid plans.353 In May 2013, AT&T, through Aio, offered prepaid customers three simple rate plans. Aio’s unlimited talk, text, and data rate plans ranged from $35 to $70 per month, with pricing varying by market. It also offered 4G download speeds of up to 4Mbs per second. Customers also had the opportunity to bring a compatible, unlocked device for activation on the Aio network (BYOD).354 In July 2013, T-Mobile launched Metro PCS in newer markets and launched a $40 rate plan that offers unlimited voice, texting and 500 MB of LTE data. Once a customer reaches 500 MB, their data speeds are slowed for the rest of their billing cycle. MetroPCS continues to offer $50 and $60 plans with larger high- speed data allotments.355 T-Mobile also expanded the MetroPCS BYOD program, which currently supports AT&T and T-Mobile iPhones, and the iPhone 5s from Verizon Wireless and Sprint. It also supports GSM- based Android phones and Windows Phone devices. T-Mobile also launched a shared Metro PCS group plan that costs $100 for 4 lines.356 Sprint and Verizon subsequently made changes in their own in-house prepaid brand offerings. The prepaid offerings included Verizon’s Allset prepaid plans for the 3G CDMA network and Sprint’s new Prepaid Smart plans. Sprint’s Boost Mobile launched an LTE promotion and then reduced prices by introducing new plans at lower price points. In May 2014, AT&T shut down the Aio brand and launched the New Cricket, which matched the Metro PCS price offers with a $40 rate plan that includes unlimited voice, texting and 500 MB of data. Like MetroPCS, New Cricket also offered a shared plan where customers can purchase four lines for $100. 160. Providers such as US Cellular and C-Spire did not launch any new prepaid rate plans during this period. For smartphones, the entry level US Cellular plan cost $50 with unlimited talk and text and 500 MB of 347 http://www.att.com/gen/press-room?pid=24185&cdvn=news&newsarticleid=36421&mapcode= 348 MetroPCS' new markets include Baltimore; Birmingham, Ala.; Cleveland, Akron, Sandusky and Toledo, Ohio; Austin, Corpus Christi, Rio Grande Valley and San Antonio, Texas; Fresno, Calif.; Houston; Memphis, Tenn.; New Orleans; San Diego, Calif.; Seattle and Tacoma, Wash.; Tallahassee, Fla.; and Washington, D.C 349 Phil Goldstein, T-Mobile expands MetroPCS foortprint by 50M POPs, adds $40 rate plan, FIERCEWIRELESS, July 25, 2013. 350 http://about.att.com/story/att_completes_acquisition_of_leap_wireless.html 351 Sue marek, AT&T revamped Cricket will take on T-Mobile’s MetroPCS with Aio-like look, rate plans, FIERCEWIRELESS, May 18, 2014. 352 See Appendix Table V.B.i for further details 353 See Appendix V.B.ii-vii for prepaid price plans. 354 http://www.att.com/gen/press-room?pid=24185&cdvn=news&newsarticleid=36421&mapcode= 355 Phil Goldstein, T-Mobile expands MetroPCS foortprint by 50M POPs, adds $40 rate plan, FIERCEWIRELESS, July 25, 2013. 356 Id 81 Federal Communications Commission DA 14-1862 data. For C-Spire the cheapest plan including data is $35 (for 30 days) and includes 350 nationwide minutes, unlimited texts, and 500 MB of data. 161. There are several MVNOs who also provide service in the prepaid space. The largest of these is TracFone, which runs on the networks of all four nationwide providers. TracFone’s Straight Talk prepaid plans offer customers unlimited voice and text, and 3 GB of high speed data for $45. When unlimited international texting is included the price increases to $60. 2. Availability of Handsets 162. One key distinction between prepaid and postpaid plans has been the differing availability of handsets. Usually, the handsets available to prepaid customers have been older models with fewer features than those available to high-end or postpaid customers, the iPhone being a good example. For instance, T-Mobile’s GoSmart brand offers lower end monthly plans with a limited phone selection, and the smartphone selection for Sprint prepaid is limited as well.357 However, smartphone availability is continuing to progress, with more options becoming available to prepaid customers.358 3. Service Coverage 163. Roaming, and hence the service coverage area, is another aspect that often distinguishes prepaid and postpaid offerings. AT&T’s prepaid GoPhone and Cricket customers are limited to service provided only on AT&T’s own network, and do not have coverage pursuant to AT&T’s roaming agreements with other providers.359 T-Mobile’s GoSmart customers have access to the T-Mobile network, but these plans do not include access to service partner networks.360 Neither Boost Mobile nor Virgin Mobile provides access to Sprint’s roaming partners. However, MetroPCS- and T-Mobile-branded prepaid plans include roaming coverage provided by service partners, and Sprint prepaid also includes the same network coverage as Sprint postpaid.361 4. Data Speeds and Data Allowances 164. Non-price data rationing continues to differentiate some prepaid smartphone offerings from postpaid offerings. Prepaid data plans typically include a monthly high-speed data allowance, after which subscribers’ data speeds are reduced. Sprint introduced greater speed reductions for heavy data users on its Boost Mobile and Virgin prepaid brands, and AT&T capped data speeds on its New Cricket brand prepaid service once a customer used all of their data allowance.362 T-Mobile offers a selection of pay-as-you-go-plans that do not include high-speed data access. T-Mobile’s GoSmart offers lower end monthly plans with no 4G data access, and employs speed reductions or service suspensions, while offering customers the option to purchase additional high- speed data. At this time, Verizon Wireless does not provide access to its LTE network with its prepaid plans. C. Price Indicators for Mobile Data 165. As the discussion above shows, there is a wide variety of pricing plans offered by the different mobile wireless providers that vary along several dimensions. As discussed earlier363 and in previous Reports, 357 http://www.sprint.com/landings/pre-paid/ 358 See Section VI.D, infra 359 http://www.att.com/maps/wireless-coverage.html#fbid=B9wW8Cw618U 360 http://www.t-mobile.com/coverage.html; http://pre-paid-phones.t-mobile.com/pre-paid-coverage; http://www.metropcs.com/metro/maps/coverage-map.jsp; https://www.gosmartmobile.com/coverage-check 361 http://www.sprint.com/landings/pre-paid/ 362 New Cricket plans slow customer speeds once their data allowance is reached. See 363 See Section II.D.1 infra 82 Federal Communications Commission DA 14-1862 it is difficult to identify sources of information that track actual mobile wireless service prices in a comprehensive and consistent manner. However, average price metrics, although imperfect, are useful in comparing trends in prices across providers and over time. Below we present two such metrics – one which shows trends in average wireless unit prices (average revenue per unit), and the other, which tracks mobile data prices (average revenue per megabyte). 1. Postpaid Smartphone Data Price 166. The Commission, as well as analysts outside the Commission, have relied on CTIA’s estimates of wireless data service revenues and, more generally, the existence of separate prices for wireless voice and data plans, to derive estimates of the unit price of wireless data service, measured in average revenue per megabyte. Without such disaggregated data it is challenging to estimate the price of wireless data service. Also, it is difficult to calculate a meaningful estimate of average revenue per megabyte actually being paid by consumers without knowing the composition of plans for each provider, the uptake rates for various plans, non-advertised promotions, and the proportion of legacy plans in a provider’s customer base. It is possible, however, to understand overall trends in smartphone data price by surveying current postpaid data offerings by mobile wireless providers. As seen in Chart V.C.1,364 the average price for data allowances above 5 GB is approximately 1 cent per megabyte or less, assuming consumers use their full data allowance. This translates into less than $10 per gigabyte of data. One caveat is that this price is calculated only for plans with data allowances and excludes unlimited data plans. The price may be lower for some heavy users on unlimited data plans. Note: Data are based on BoA/ML, “A Frantic Start to 2014 in Wireless Pricing”, 4 April 2014, and company websites. Only unlimited talk + text plans are included. Unlimited data plans are excluded. 364 The data are presented in Appendix Table V.C.i. 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 $/ M B Data Allowance Chart V.C.1 Post-Paid Smartphone Data Price ($/MB) Unlimited Talk+Text Plans 1Q 2014 AT&T Verizon Wireless Sprint T Mobile T-Mobile and Sprint's unlimited talk and text plans mostly have unlimited data as well, rather than tiered data allowances. 83 Federal Communications Commission DA 14-1862 2. Prepaid Smartphone Data Price 167. Chart V.C.2 shows the current price per megabyte ($/MB) of prepaid smartphone data for the four national providers and their associated MVNOs. Data allotments included with prepaid plans are priced slightly higher than data allotments included with postpaid plans. There is a cluster of prepaid smartphone plans with high-speed data allowances of 500 MB and a data price of approximately eight cents per megabyte. This translates into roughly $8 per GB of data, assuming a consumer uses the maximum allowed by the price plan. As data allowances increase, price per megabyte drops considerably. For example, plans with a 5 GB data bucket cost approximately 1.5 cents per megabyte for customers consuming their full data allowance. Sprint (along with its MVNOs such as Boost and Virgin) and also T-Mobile (along with MetroPCS) offer plans that span both the high and low end of the data buckets.365 Note: Data are based on Bank of America analysts’ reports and staff calculations. Only unlimited talk + text plans are included. The AT&T plans include Cricket and GoPhone plans, and those of T-Mobile include the provider’s own prepaid plans and GoSmart plans. VI. NON-PRICE RIVALRY 168. Mobile wireless service providers also compete for customers on dimensions other than price, including investment, network coverage and technology, service quality, as well as other factors such as advertising and marketing. Providers take actions and make expenditures to differentiate themselves from competitors and to imitate initiatives of their competitors that have been successful in attracting customers. Such non-price rivalry can influence a customer’s choice of a provider and impose significant competitive constraints, especially in high technology industries that experience rapid innovation. This section presents data in five broad categories reflecting non-price rivalry among mobile wireless service providers: investment; network coverage and technology upgrades; quality of service; differentiation in handsets and devices; and other factors such as advertising and marketing. 365 Prepaid pricing data are listed in more detail in Appendix Tables V.B.iii-vii. 0.00 0.02 0.04 0.06 0.08 0.10 500 MB 1 GB 2.5 GB 3 GB 5 GB $/ M B Data Allowance Chart V.C.2 Pre-paid Smartphone Data Price ($/MB) 1Q 2014 AT&T/GoPhone Sprint Virgin Mobile Verizon Wireless T-Mobile Boost Mobile MetroPCS 84 Federal Communications Commission DA 14-1862 A. Investment 169. Mobile wireless service providers differentiate themselves in the marketplace by improving the customer network experience through improvements in capacity, coverage, and service quality. Providers have been able to expand into new geographic areas and/or upgrade networks in existing markets after adding to their spectrum portfolios through participation in spectrum auctions and secondary market transactions. Providers have also expanded their network coverage and capacity through increased investment in and expansion of their existing assets and infrastructure. In this section, we focus on non-spectrum-related investment, which is one of the ways in which wireless mobile providers compete in the marketplace. Some providers make strategic capital expenditure (CAPEX) decisions to differentiate their service offerings from those of rivals by becoming the first to deploy a particular upgrade or new network technology. Other providers wait for rivals to make the first move and then respond by upgrading their own networks.366 170. Wireless providers in the U.S. have spent more than $134 billion in capital investments during the past five years.367 Incremental capital investment by wireless providers rose to $33.1 billion in 2013, a 10.1 percent increase from the $30.1 billion spent in 2012. Verizon Wireless, AT&T, Sprint, and T-Mobile spent a combined $16 billion in the first half of 2014 and $31.5 billion in 2013, accounting for more than 96 percent of total industry capital investment in these time periods.368 AT&T and Verizon Wireless together spent $11.9 billion in the first half of 2014, over 71 percent of the industry total. 369 This had spent $20.6 billion on capital investment in 2013, which was over 63 percent of the industry total.370 Chart VI.A.1 below shows the capital expenditures for the four national providers, as well as for selected regional providers, during the past few quarters. As seen from the chart, capital expenditures have continued to vary significantly amongst providers. AT&T and Verizon Wireless continued to invest more than Sprint or T-Mobile by wide margins. Apart from a temporary increase in 2Q12, neither AT&T nor Verizon Wireless significantly increased its capital expenditures between 2Q11 and 2Q14. For Sprint and T-Mobile, the 4Q12 increase appears to be more persistent, and both providers show higher levels of capital expenditure in 4Q13 compared to 2Q11. In its comments,371 Verizon Wireless states that since 2000, it has invested more than $80 billion in its network, with capital expenditures of more than $26 billion in the last three years alone. 366 See Sixteenth Report, 28 FCC Rcd at 3836 at ¶ 219 367 CTIA Year-End 2013 Wireless Indices Report, at 96. CTIA’s figure includes incremental investment in currently operational systems, including expenditures for building operating systems, land and capital leases, and all tangible non- system capital investment, but does not include the cost of spectrum licenses purchased at auctions or other acquisition processes or greenfield builds. 368 UBS 411 Report, Version 51, April 2014. UBS 2Q 2014. T-Mobile includes MetroPCS 369 Id 370 Id 371 See Verizon Wireless Comments at 26 85 Federal Communications Commission DA 14-1862 Notes: Based on UBS Wireless 411 Report, Version 51 at 28. UBS 411 2Q 2014. Data in the chart is for second and fourth quarter. Metro PCS data are not available separately after the fourth quarter of 2012 as the T-Mobile and MetroPCS merger was consummated in early 2013. Leap is reported separately from AT&T as the AT&T and Leap merger was not consummated by the fourth quarter of 2013. 171. Looking beyond the short-term data in Chart VI.A.1, we see that an increase in capital expenditures has taken place over the last six years for the national providers.372 In Chart VI.A.2 below we present annual capital expenditures for the four nationwide providers from 2009 – 21st half 2014.373 AT&T steadily increased its nominal investment. Sprint more than doubled its investment from 2011– 2013. Capital expenditures by Verizon Wireless and T-Mobile have held fairly steady from 2010 – 2013, with a slight increase in 2013. However, there appears to be substantial variation in both the level and growth of CAPEX, even amongst national providers. Source: Company SEC 10-K filings and UBS Wireless 411, Version 51, UBS 411 Version 54. 372 For more details, see Appendix Table VI.A.i 373 US Wireless 411 Version 51, March 2014 0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 1Q11 2Q11 3Q11 4Q11 1Q12 2Q12 3Q12 4Q12 1Q13 2Q13 3Q13 4Q13 1Q14 2Q14C ap E x ($ in m il li on s) Quarters Chart VI.A.1 Quarterly Capital Expenditure by U.S. Mobile Wireless Providers 2011-Ist Half 2014 AT&T VZW Sprint T-Mobile US Cellular Metro PCS Leap Other Providers $0 $5,000 $10,000 $15,000 2009 2010 2011 2012 2013 1st Half 2014 C ap it al E xp en di tu re s ($ M il li on s) Chart VI.A.2 Yearly Capital Expenditures by Service Provider 2009 - 1st Half 2014 Verizon Wireless AT&T Sprint T-Mobile 86 Federal Communications Commission DA 14-1862 172. Variations in CAPEX may not be synchronized across providers for several reasons. First, providers follow different technological migration paths, which may be on different timeframes. Recently, the industry has followed distinct technological migration paths for LTE upgrades, with each provider implementing its own sequence of upgrades. As a result, CAPEX can vary from one service provider to the next. Second, providers often base their investment decisions on an assessment of how network deployments and upgrades affect future earnings. Third, the timing of network investments often has a strategic component vis-à-vis rivals, as discussed above. Finally, access to capital may be difficult for some providers, and this may hinder investment. According to NTCA, which consists exclusively of small, rural providers, 68 percent of the rural providers who were surveyed described the process of obtaining financing for their wireless projects as “fairly difficult” or “very difficult”, while another 13 percent found it “virtually impossible”.374 B. Network Coverage and Technology Upgrades 173. Network investment remains a centerpiece of service providers’ efforts to improve their customers’ mobile wireless service experience. During 2013 and early 2014 several providers continued to upgrade and expand their networks with technologies that enable faster data transfer speeds. Other providers announced plans to make additional upgrades in the near future.375 As discussed below, a critical way in which mobile wireless service providers differentiate themselves is with the speeds, reliability, capabilities, and coverage of their mobile broadband networks.376 Most wireless providers offer national coverage, using a combination of their own facilities and roaming arrangements. Since coverage and performance remain key elements of competition, small, regional, and national providers alike continue to invest substantially in their networks.377 In the following discussion, we consider current network coverage by provider, technology and roaming by provider, and future network deployment plans by provider, including the implications for competition in the mobile wireless industry. 1. Current Coverage by Provider 174. This section presents an overview of wireless voice and broadband coverage by provider. As discussed earlier,378 for purposes of this Report , mobile wireless coverage represents either mobile voice or mobile broadband coverage, and “mobile broadband” includes coverage and services offered using the following 3G and 4G technologies: EVDO, EVDO Rev A, WCDMA/HSPA, HSPA+, LTE, and mobile WiMAX. 175. Similar to the analysis of nationwide mobile coverage in Chapter III, the discussion in this section is based on U.S. census blocks overlaid on coverage maps provided to the Commission through a contract with Mosaik Solutions. As discussed earlier, these coverage estimates represent deployment of mobile networks and do not indicate the extent to which providers actually offer service to any or all residents in the covered areas. While recognizing that this analysis likely overstates the coverage experienced by consumers because of limitations in Mosaik data, we find that this analysis is useful because it provides a general baseline that can be compared over time across network technologies, and providers.379 a. Mobile Wireless Network Coverage by Provider 176. Mobile voice coverage by provider is presented in Chart VI.B.1.380 Each of the four nationwide 374 NCTA, 2013 Wireless Survey Report, January 2014, at 3 and 10. 375 See Table VI.B.1 and Section VI.B.3, infra. 376 See Sixteenth Report, 28 FCC Rcd at 3836 at ¶ 182; AT&T Comments at 31; WCAI Reply at 5. 377 See Verizon Wireless Comments at 27 378 See Section III.A infra 379 See Section II.A infra 380 Also see Appendix Tables VI.B.i and VI.B.ii 87 Federal Communications Commission DA 14-1862 providers covered census blocks containing at least 90 percent of the population with their respective mobile wireless networks as of January 2014. Verizon Wireless and AT&T each covered approximately 85 percent of U.S. road miles, while Sprint covered 48.1 percent and T-Mobile covered 58.3 percent of U.S. road miles. Verizon Wireless and AT&T each covered approximately 61 percent of U.S. land area with their respective mobile wireless networks, while Sprint and T-Mobile each covered less than 35 percent of land area. Source: Staff calculations based on January 2014 Mosaik coverage data and 2010 Census population data. We note the data underlying these estimates measure mobile network “coverage,” and not network providers affirmatively offering service to residents in all those locations. Also, calculations based on Mosaik data have certain limitations that likely overstate the extent of mobile wireless coverage. b. Mobile Broadband Network Coverage by Provider 177. Mobile broadband coverage is presented in Chart VI.B.2.381 Verizon Wireless and AT&T each covered census blocks containing at least 97 percent of the population with mobile broadband as of January 2014, while Sprint covered 89.3 percent and T-Mobile covered 78.8 percent.382 Verizon Wireless and AT&T each covered over 79 percent of US road miles and over 55 percent of U.S. land area, while Sprint and T-Mobile each covered less than 50 percent of US road miles and less than 25 percent of US land area with mobile broadband. 381 Also see Appendix Tables VI.B.iii and VI.B.iv 382 Commission estimates based on census block analysis of Mosaik Coverage Right coverage maps, January 2014.The estimates for the regional providers apply both to mobile wireless coverage and to mobile broadband coverage. 34.3% 24.9% 61.8% 61.2% 58.3% 48.1% 85.3% 84.8% 92.1% 89.8% 98.7% 96.9% 0.0% 20.0% 40.0% 60.0% 80.0% 100.0% T-Mobile Sprint ATT Verizon Wireless Mobile Wireless Voice Coverage (%) Chart VI.B.1 Estimated Mobile Wireless Coverage in the U.S. by Provider , January 2014 % of U.S. Population % of U.S. Road Miles % of U.S. Square Miles 88 Federal Communications Commission DA 14-1862 Source: Staff calculations based on January 2014 Mosaik coverage data and 2010 Census data on population. We note the data underlying these estimates measure mobile network “coverage,” and not network providers affirmatively offering service to residents in all those locations. Also, calculations based on Mosaik data have certain limitations that likely overstate the extent of mobile broadband coverage. c. Urban/Rural Comparisons (i) Mobile Wireless Network Coverage 178. As seen in Chart VI.B.3, each of the four nationwide providers covered census blocks containing at least 95 percent of the non-rural population with their mobile wireless voice networks, as of January 2014.383 In terms of the rural population, Verizon Wireless and AT&T covered census blocks containing 89.1 percent and 91.2 percent, respectively, while Sprint and T-Mobile covered census containing 57.8 and 65.9 percent, respectively. Each of the four nationwide providers covered a significantly higher percentage of non-rural than rural land area and road miles. Source: Staff calculations based on January 2014 Mosaik coverage data and 2010 Census data on population. We note the data underlying these estimates measure mobile network “coverage,” and not network providers affirmatively offering service to residents in all those locations. Also, calculations based on Mosaik data have certain limitations that likely overstate the extent of mobile broadband coverage. 383 See Appendix Tables VI.B.v and VI.B.vi for more detailed data on estimated mobile wireless coverage by provider in rural areas and non-rural areas, respectively. 15.0% 24.5% 55.1% 59.5% 32.3% 47.4% 79.2% 83.6% 78.8% 89.3% 97.8% 96.9% 0.0% 20.0% 40.0% 60.0% 80.0% 100.0% T-Mobile Sprint ATT Verizon Wireless Mobile Wireless Broadband Coverage (%) Chart.VI.B.2 Estimated Mobile Wireless Broadband Coverage in the U.S. by Provider January 2014 % of U.S. Population % of U.S. Road Miles % of U.S. Square Miles 65.9% 57.8% 91.2% 89.1% 97.8% 95.1% 99.8% 97.9% 0.0% 20.0% 40.0% 60.0% 80.0% 100.0% T-Mobile Sprint ATT Verizon Wireless Mobile Wireless Coverage (%) Chart VI.B.3 Estimated Mobile Wireless Voice Coverage in Non-Rural and Rural Areas by Provider; January 2014 % of Non- Rural U.S. Population % of Rural U.S. Population 89 Federal Communications Commission DA 14-1862 (ii) Mobile Broadband Network Coverage 179. As seen in Chart VI.B.4, each of the four nationwide providers covered census blocks containing at least 90 percent of the non-rural population with their mobile wireless broadband networks, as of January 2014.384 In terms of the rural population, Verizon Wireless and AT&T covered census blocks containing 91.2 percent and 89.7 percent, respectively, while Sprint covered census blocks containing 59.6 percent and T-Mobile covered census blocks containing 29.4 percent of the rural population. Each of the four nationwide providers covered a significantly higher percentage of non-rural than rural land area and road miles. Source: Staff calculations based on January 2014 Mosaik coverage data and 2010 Census data on population. We note the data underlying these estimates measure mobile network “coverage,” and not network providers affirmatively offering service to residents in all those locations. Also, calculations based on Mosaik data have certain limitations that likely overstate the extent of mobile broadband coverage. 2. Coverage and Roaming 180. Service providers may use roaming services to enhance their coverage for a variety of reasons, including temporary arrangements while their networks are being deployed, and as permanent arrangements due to the economics of the market or to their business models, as discussed earlier in Section IV.C. No facilities- based provider – including the four nationwide providers – has built out its entire licensed service area, and consequently all employ roaming to some extent to fill gaps in their coverage.385 In addition, there are non- nationwide providers whose business plans do not include nationwide networks. Many of these non-nationwide providers are able to offer their customers coverage that is national in scope through roaming agreements with other mobile wireless providers. Accordingly, roaming remains particularly important for small and regional providers, allowing them to compete with nationwide providers for customers in their network service areas.386 Similarly, roaming provides important assistance to potential new entrants who wish to begin offering service 384 See Appendix Tables VII.B.vii and VII.B.viii for more detailed data on estimated mobile wireless broadband coverage by provider in rural areas and non-rural areas, respectively. 385Sixteenth Report, 28 FCC Rcd at 3836 ¶ 208; Roaming Order on Reconsideration, 25 FCC Rcd at 4192 ¶ 23. One potential measure of the significance of roaming in the wireless industry is roaming revenues, which are discussed in detail below. See also AT&T Reply Comments at 18-19. 386Sixteenth Report, 28 FCC Rcd at 3836 ¶ 208; see also RCA Comments at 15; NTCA Comments at 3-4 (arguing that regional and local providers offer a small footprint and need to partner with other providers through roaming agreement to offer their subscribers competitive expanded coverage.). 29.4% 59.6% 89.7% 91.2% 90.4% 96.2% 99.7% 98.2% 0.0% 20.0% 40.0% 60.0% 80.0% 100.0% T-Mobile Sprint ATT Verizon Mobile Broadband Coverage (%) Chart VI.B.4 Estimated Mobile Broadband Coverage in Non-Rural and Rural Areas, by Provider; January 2014 % of Non- Rural U.S. Population % of Rural U.S. Population 90 Federal Communications Commission DA 14-1862 before they have fully deployed their networks.387 181. By definition, MVNOs and resellers rely on using the networks of one or several facilities-based providers to compete with the nationwide providers. As they do not have networks of their own, it is essential that they maintain wholesale arrangements with facilities-based service providers in order to provide coverage. Depending on the particular arrangement, customers of MVNOs and resellers may have limited or no access to networks other than that of the underlying wholesale provider For example, the coverage experienced by customers of TracFone’s Straight Talk388 varies depending on the underlying wholesale provider. A Straight Talk customer obtaining service on an underlying CDMA network, for instance, will not have access to voice and SMS service when not within the coverage of that network. A Straight Talk customer whose service is provided on an underlying GSM network, by contrast, may have voice and SMS service when outside that network’s coverage area. 389 3. Service Provider Network Deployments 182. Network investment remains a centerpiece of service providers’ efforts to improve their customers’ mobile wireless service experience. During the past few years, several providers upgraded and expanded their networks with technologies that enable faster data transfer speeds. Other providers announced plans to make additional upgrades in the near future.390 While service providers initially upgraded their networks with various technologies, all of the major mobile wireless providers now offer or plan to deploy LTE. The following section includes a brief discussion of the deployment strategies for each of the top five providers, as well as a detailed discussion of the mobile network upgrades of the major mobile wireless providers. For purposes of this Report, we include all 3G (CDMA EV-DO, EV-DO Rev. A, WCDMA/UMTS/HSPA, HSPA+) and 4G (LTE, and mobile WiMAX) network technologies in our discussion of mobile broadband.391 While the Mosaik deployment data distinguish among different mobile wireless network technologies, other factors than network technology may affect network performance. These factors may include the configuration of the network, the amount of spectrum used, and the type and capacity of backhaul connection to the cell site.392 Below, we provide a detailed look at network deployment plans by provider and include a detailed discussion of deployment plans. 387Sixteenth Report, 28 FCC Rcd at 3836 ¶ 208; see also Roaming Order on Reconsideration, 25 FCC Rcd at 4191-92 ¶ 21 (recognizing that without the ability to offer roaming in markets where they hold spectrum, new entrants would in effect be required “to build out their networks extensively throughout the newly obtained license area before they can provide a competitive service to consumers, all without the benefit of financing the construction of new networks over time with revenues from existing services and reliance on roaming to fill in gaps during build out”); see also NTCA Comments at 3. 388 http://www.straighttalk.com/wps/portal/home/h/about (visited July 7, 2014) 389 http://www.straighttalk.com/wps/portal/home/h/legal/terms-and-conditions (visited July 7, 2014) 390 See Section IV.B.1.A, infra. 391 The terms “3G” and “4G” are used by industry for marketing purposes, as well as by the International Telecommunications Union (ITU) for technical specifications. See Sixteenth Report, 28 FCC Rcd at 3796 ¶ 186 392 2012 Eighth Broadband Progress Report, GN Docket No. 11-121, (rel. Aug. 21, 2012), ¶ 40. 91 Federal Communications Commission DA 14-1862 Table VI.B.1 3G/4G/ LTE Deployment Reported by Selected Mobile Wireless Service Providers As of December 2014 Service Provider HSPA, HSPA+, and EV-DO Deployment LTE and WiMAX Deployment Additional Discussion AT&T Wireless Entire network is covered by HSPA+, covering 300 million POPs. AT&T covered 320 million393 POPS with 4G/LTE as of 4Q14, up from 300 million POPs in 2Q14, 280 million POPs in 4Q13 and approximately 250 million POPs in 3Q13.394 AT&T highlighted that about 50 percent of postpaid smartphones are now LTE compatible. Verizon Wireless EV-DO Rev. A network covered 300 million POPs. As of 4Q14 Verizon Wireless covered 308 million POPs in over 500 US markets with LTE, accounting for 98 percent of POPs.395 Verizon’s LTE network over layed 99 percent of its 3G EV- DO mobile broadband network. XLTE, is now available in more than 400 markets across the country. Verizon had already migrated 54 percent of its data traffic to its LTE network. It is now adding capacity to its 4G LTE network using AWS spectrum. The additional bandwidth is called XLTE Verizon Wireless – LTE in Rural America Partners 13 providers had launched LTE and covered 1.8 million POPs.396 Program included 20 small, rural providers that had already launched or plan to launch LTE to areas covering approximately 2.8 million people across 14 states. Sprint EV-DO Rev. A network covered approximately 277 million POPs. Sprint’s LTE and WiMAX deployments covered 250 million POPs397 in 470 markets as of 4Q14, up from 200 million POPs in 443 markets as of 4Q13, and 230 markets in 3Q13.398 Sprint expects to expand its coverage to over 2.5 million square miles399 through its partnership with the Competitive Carrier’s Association Data Roaming Sprint continues to target 250 million 4G / LTE covered POPs by year end 2014 under the Network Vision project, using FDD- LTE with 800 / 1900 MHz spectrum. 393 http://www.att.com/network/en/index.html (accessed December 9, 2014) 394 http://about.att.com/content/dam/snrdocs/4g_evolution_infographic.pdf (accessed Mar. 26, 2014) 395 http://www.verizonwireless.com/news/LTE/Overview.html (accessed December 9, 2014 and Mar. 25, 2014) 396 https://aboutus.verizonwireless.com/technology/network/ (accessed December 9, 2014 and Mar. 25, 2014) 397 http://shop.sprint.com/modals/4g_lte_plan_details.html (accessed December 9, 2014) 398 http://www.sprint.com/netdotcom/index.html (accessed December 9, 2014 and March 24, 2014) 399 http://newsroom.sprint.com/news-releases/sprint-competitive-carriers-association-and-netamerica-alliance-join-forces-to- accelerate-deployment-and-utilization-of-4g-lte-across-the-united-states.htm (accessed March 27 2014) 92 Federal Communications Commission DA 14-1862 Hub. Sprint also formed a strategic partnership with NetAmerica Alliance to expand 4G LTE coverage in rural markets. T-Mobile HSPA+ 21 network covered over 200 million POPs and HSPA+ 42 network covered 184 million POPs. As of 4Q14, T-Mobile's 4G LTE network reached 273 metro areas nationwide. It covered 230 million people in 4Q14, up from 210 million people in 1Q14.400 T-Mobile is the process of upgrading its 2G/EDGE network to 4G/LTE, with plans to complete 50 percent of the work in 2014, and to substantially complete the upgrade by the middle of 2015. 183. AT&T: In its Comments, AT&T stated that as of April 2013 it had built out LTE to cover more than 200 million POPs, and it expected to reach 90 percent of its planned 300 million POP LTE deployment by the end of 2013.401 As of December 2014, its LTE network covered 320 million POPS, as seen in Table VI.B.1 above. AT&T has announced plans to deploy commercial mobile broadband services using carrier aggregation technology - which is part of the LTE Advanced specifications - to combine transmissions across either AWS or PCS high-band band spectrum with 700 MHz D and E block spectrum.402 AT&T is also exploring the possibility of offering eMBMS403 services on these bands.404 AT&T is developing a "broadcast capability" to remove video traffic from its wide-area wireless networks.405 . AT&T is also exploring using LTE Advanced technology to ensure that it can meet users' data demands as more customers start using LTE.406 AT&T launched Voice over LTE (VoLTE) in select markets in May 2014.407 184. Verizon: Verizon Wireless augmented its LTE network in 50 different cities with AWS spectrum in the first half of 2014 to avoid potential capacity issues, as more than 66 percent of the company’s data traffic now rides on 4G LTE.408 In December 2014, it covered 308 million POPs as seen in the table above. The company will begin to re-farm PCS spectrum to LTE from its 3G network in 2015.409 In every major city east of the Mississippi and in several western markets, Verizon Wireless is using 40 megahertz of spectrum, compared to the 20 megahertz it has deployed on its 700 MHz Upper C Block spectrum for its macro LTE deployment.410 In 400 http://www.t-mobile.com/coverage.html (accessed December 9, 2014 and March 24, 2014) 401 See AT&T Comments at 9 402 http://apps.fcc.gov/ecfs/comment/view?id=6017610610 403 eMBMS refers to evolved multimedia broadcast/ multicast service on LTE advanced, see http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=6353684&tag=1 404 “”AT&T ‘exploring the possibility” of LTE Broadcast with eMBMS”, Mike Dano, Fierce Wireless, April 2, 2014, http://www.fiercewireless.com/story/att-exploring-possibility-lte-broadcast-embms/2014-04-02 405 http://www.fiercewireless.com/story/att-use-lower-700-mhz-d-and-e-block-spectrum-lte-broadcast/2013-09- 24#ixzz2yIw98jwV 406 http://www.fiercewireless.com/story/att-edging-lte-advanced-technologies-capacity-not-speed/2014-02- 26#ixzz2yJ0Mj5pX 407 http://about.att.com/story/att_introduces_high_definition_voice_in_initial_markets.html 408 http://www.verizonwireless.com/news/article/2013/12/verizon-wireless-4g-lte-three-year-anniversary.html 409 http://www.fiercewireless.com/story/verizon-start-refarming-pcs-spectrum-lte-2015/2013-06-27 410 http://www.fiercewireless.com/story/verizon-will-deploy-aws-spectrum-50-markets-mid-2014/2013-12- 09#ixzz2yJ2KLBIM 93 Federal Communications Commission DA 14-1862 addition, after several delays, 411 in September 2014, Verizon Wireless has started rolling out VoLTE service nationwide but is initially making VoLTE available on only two smartphones.412 4G LTE roaming for Verizon Wireless customers outside the U.S. will also begin in select countries.413 Verizon Wireless’ LTE in Rural America (LRA) program allows Verizon Wireless to expand its 4G LTE network into rural areas, and to allow customers of participating companies to roam on Verizon Wireless’ 4G LTE network throughout the U.S., including Alaska.414 185. Sprint: Sprint is in the process of replacing its WiMAX technology with LTE. The transition is expected to be complete by the end of 2015, at which point the WiMAX network will be completely decommissioned.415 Sprint also shut down its iDEN Nextel network in 2013.416 Once WiMAX is transitioned off of Sprint’s 800 megahertz band in a given market, this spectrum will be used to launch Sprint Spark. Sprint Spark harnesses three different frequencies in the LTE spectrum, and actively cycles between them depending on usage and need.417 In its Comments,418 Sprint states that one important facet to Sprint’s competitive efforts has been its Network Vision project, which is an initiative to consolidate Sprint’s networks and technologies into a single nationwide 3G and 4G network. .Sprint 4G LTE service is available to 250 million Americans in 470 markets in December 2014, and Sprint expects 100 million Americans will have Sprint Spark or 2.5GHz coverage by the end of this year as well.419 By December 2014, Sprint Spark was live in 20 cities with plans for expanding to 100 more cities in the next three years.420 Sprint plans to eventually launch VoLTE, but there is currently no established timeline.421 186. In addition, Sprint has announced that it will partner with both CCA and NetAmerica to accelerate the deployment of 4G LTE in rural communities.422 Through these partnerships, CCA providers and NetAmerica Alliance Members are positioned to more efficiently and quickly deploy and support their owned and operated 4G LTE networks. In turn, Sprint’s customers will be able to roam on the 4G LTE networks being built by CCA and NetAmerica Alliance Members across the country. To facilitate roaming between these networks, beginning in January 2015, Sprint plans to offer 4G LTE devices that will include a chipset allowing the devices to roam on the lower 700 MHz spectrum primarily in use by CCA and NetAmerica Alliance Members, which is 411 http://www.fiercewireless.com/story/att-admits-volte-delay-wont-offer-new-launch-date/2014-02-26; http://www.forbes.com/sites/greatspeculations/2014/08/20/verizon-to-launch-volte-service-by-december-delays-launch-of- lte-only-phones/ 412 “Verizon starts rolling out VoLTE, but on only 2 phones to start”, Phil Goldstein, Fierce Wireless, September 17, 2014; http://www.fiercewireless.com/story/verizon-starts-rolling-out-volte-only-2-phones-start/2014-09-17 413 http://www.verizonwireless.com/news/article/2013/12/verizon-wireless-4g-lte-three-year-anniversary.html 414 http://www.verizonwireless.com/news/2013/09/4g-lte-rural-america-program-alaska.html 415 http://www.sec.gov/Archives/edgar/data/101830/000010183014000012/sprintcorp201310-k.htm 416 http://www.fiercewireless.com/story/sprint-shutter-wimax-network-end-2015-will-turn-least-6000-clearwire-sites/2014- 04-07#ixzz2yQUccesC 417 http://faster.sprint.com/2014/03/17/sprint-spark-arrives-in-2-new-markets/?INTMKT=MA:MS:103013: SparkHub:Articles_RelatedContent 418 See Sprint Comments at 7 419 http://newsroom.sprint.com/presskits/sprint-spark.htm, (accessed December 8, 2014 and September 10, 2014) 420 http://www.sprint.com/netdotcom/ (accesed December 9, 2014) 421 http://www.broadsoft.com/news/2014/sprint-selects-broadsoft-to-deliver-next-generation-ims-and-voice-over-lte-services/ 422 http://newsroom.sprint.com/news-releases/sprint-competitive-carriers-association-and-netamerica-alliance-join-forces-to- accelerate-deployment-and-utilization-of-4g-lte-across-the-united-states.htm 94 Federal Communications Commission DA 14-1862 notable because Sprint doesn't own any 700 MHz spectrum.423 187. T-Mobile: In 2013 T-Mobile made substantial investments to upgrade to LTE.424 During the first quarter of 2013 alone, T-Mobile invested $1.1 billion, in support of an accelerating network modernization program. 425 T-Mobile’s LTE coverage now reaches 250 milion people, up from 220 million POPs in earky 2014. 426 Where it does not offer 4G LTE, T-Mobile customer devices will automatically transition to its 4G HSPA+ network. Under T-Mobile’s network strategy, 4G HSPA+ essentially serves as a fallback such that consumers can access 4G coverage with multiple technologies. T-Mobile has continued to launch its LTE network, and is now kicking off a new program to upgrade its 2G/EDGE network with 4G LTE. The company plans to complete 50 percent of the work in 2014, and expects the program to be substantially complete by the middle of 2015.427 The upgrade will provide customers who currently experience 2G/EDGE coverage new access to 4G LTE, and will also expand the existing 4G LTE network. T-Mobile is in the process of acquiring 700 MHz A-Block spectrum, on which they plan to deploy 4G LTE.428 T-Mobile has launched VoLTE across its 4G LTE network.429 T-Mobile is now known as T-Mobile USA, after a merger with MetroPCS. MetroPCS is marketed as a separate brand. The MetroPCS footprint has continued to expand, first through the Apollo 15 program, and soon through the upcoming Apollo 30 program. 188. U.S. Cellular and Other Providers: U.S. Cellular plans to add more than 1,200 4G LTE cell sites in 2014, and to expand the existing 4G LTE service in 13 states. By the end of 2014, more than 93 percent of U.S. Cellular customers will have access to 4G LTE.430 In 2013, U.S. Cellular sold customers and certain PCS license spectrum in the Chicago, central Illinois, St. Louis and certain Indiana/Michigan/Ohio markets to Sprint.431 They then focused their LTE expansion in the remaining markets. In addition to the providers discussed above, several other smaller, regional operators had deployed 3G and 4G technologies within their networks as of January 2014.432 C. Quality of Service 189. Key characteristics for mobile wireless performance include network speeds, latency and packet loss. The Commission has recognized the importance of accurate and timely data on these characteristics in informing consumer decisions, Commission policy, and service provider network investment decisions. This 423 http://www.fiercewireless.com/story/sprint-add-700-mhz-band-12-capabilities-some-new-devices-starting-next-year/2014- 03-26#ixzz2yQMeKWmI 424 See AT&T Comments at 10; T-Mobile, Press Release, T-Mobile USA Reports First Quarter 2013 Results (Mar. 8, 2013), available at http://investor.t-mobile.com/phoenix.zhtml?c=177745&p=irol-IRHome (“T-Mobile USA 2013 First Quarter Results”).; Phil Goldstein, T-Mobile boasts of lead in LTE Advanced, FierceWireless, Apr. 23, 2013, available at http://www.fiercewireless.com/story/t-mobile-boasts-lead-lte-advanced/2013-04-23 . 425 Se T-Mobile Comments at 20-21. 426 See Table VI.B.1 infra. http://newsroom.t-mobile.com/issues-insights-blog/the-un-carrier-network-designed-data- strong.htm?AID=11031750&PID=6147683&SID=ovrfcgqnm36a; http://www.fiercewireless.com/tech/story/t-mobiles-data- strong-network-gains-lte-coverage-bandwidth/2014-06-19 427 http://newsroom.t-mobile.com/news/t-mobile-celebrates-1st-anniversary-of-lte-rollout-by-launching-major-network- upgrade-program.htm 428 http://newsroom.t-mobile.com/phoenix.zhtml?c=251624&p=irol-newsArticle&ID=1908666&highlight= 429 http://newsroom.t-mobile.com/issues-insights-blog/firing-on-all-cylinders-earnings-jdp.htm 430 http://www.uscellular.com/about/press-room/2014/USCellular-Announces-New-Markets-to-Receive-4G-LTE-Service-in- 2014.html 431 http://usc.q4cdn.com/bd464866-e7e0-4821-8879-8a5c2a35f568.pdf 432 Infra 95 Federal Communications Commission DA 14-1862 Report will primarily analyze speed data using the Ookla Net Index data, data from the FCC Speed Test App, RootMetrics data, and the CalSPEED drive-test data gathered by the California Public Utility Commission. 190. Mobile broadband network service quality experienced by consumers may vary greatly with a number of real world factors such as the service provider’s received signal quality, cell traffic loading and network capacity in different locations, as well as the capability of consumers’ devices.433 First, mobile connection quality will vary based on the location of the receiving device in reference to the transmitting device, which is often a cellular tower. If the receiving device (and the person using it) is behind a wall, blocked by terrain or otherwise has an impaired connection with the tower, the mobile broadband service will be degraded or not available. Second, the performance of the broadband connection degrades over distance to the tower, even with a clear line of sight. Performance at the edge of a tower’s coverage is not equal to performance close to the tower. Third, cellular signals are shared by many users—the more simultaneous usage, the lower the potential performance of any one connection.434 It is also important to note that for all mobile technologies, speed and performance measurements are only valid when a wireless connection can be accessed. “Dead zones” and loss of signal reduce wireless effectiveness.435 Moreover, from the customer’s perspective, overall network performance is the product of more than network quality alone and often reflects differences in device capability as well.436 For data services, network quality as perceived by the customer may also be use-, case-, or application-dependent (e.g., a consumer who solely uses e-mail may view the quality of the network differently than one who streams video regularly). Furthermore, consumers may place more weight on one particular aspect of network quality than another – such as coverage or peak data speeds – when choosing their mobile wireless services.437 191. In recognition of the effects of these different parameters on mobile network performance, mobile network speeds are commonly assessed using various methodologies. The two most prevalent approaches rely on crowdsourced data or drive-test data. Crowdsourced data are user-generated data produced by consumers who voluntarily download speed test applications on their mobile devices. These apps commonly collect data on the provider, location of device, download and upload speeds, latency and packet loss, which are then transmitted to the company or entity that developed the app. In some cases, the apps automatically schedule these tests to run at certain times during the day, while in others, the user has to choose to run the tests. Generally, crowdsourced data can bring the benefits of generating a large volume of data at a very low cost and of measuring actual consumer experience on a network in a wide variety of locations, indoor and outdoor. We note, however, that crowdsourced data are often not collected pursuant to statistical sampling techniques, and may require adjustments to construct a representative sample from the raw data. For instance, crowdsourced mobile data come from a self-selected group of users, and there often is little control for most tests regarding such parameters as when people implement the test, whether the test is performed indoors or outdoors, the geographic location of the tester, and the vintage of 433 For example, the received signal quality is dependent on the service provider’s deployed cell site density, low/high frequency radio wave propagation losses, user locations, indoor obstructions and outdoor foliage or clutter, weather, inter-cell interference conditions, and wireless network optimization parameters. The cell traffic loading or demand is dependent on the overall number of concurrent active mobile broadband users sharing the same cell, which in turn depends on user locations, the day of the week, and the time of the day. The capacity of a provider’s wireless network is dependent on the deployed mobile wireless technology, sites and equipment, available bandwidth, and enhanced backhaul connections. See Sixteenth Report at ¶ 290. 434 The FCC Omnibus Broadband Initiative (OBI). Broadband Performance - OBI Technical Paper No. 4. at 19 435 Id at 19-20 436 The capability of consumer devices (e.g. smartphones, tablets, USB dongles, and laptops) could result in users experiencing different data speeds on the same mobile wireless broadband network. Even differing capabilities within each device category, such as smartphone processing power and memory, could result in better user experiences on 4G networks. 437 See Consumer Satisfaction with Service Providers, infra, for a discussion of overall consumer satisfaction with their mobile wireless services. 96 Federal Communications Commission DA 14-1862 the consumer’s device.438 Drive test data, by contrast, are generated from tests that control for the location and time of the tests as well as for the devices. Drive tests, however, are more expensive to conduct, involve significant judgment about when and where the tests are run, often do not involve significant testing indoors or in many rural areas, and typically produce datasets that are not as rich as crowdsourced data – all of which are likely to have some effects on reported results. 192. Ookla is one of the most prominent providers of crowdsourced data. The FCC has also made available a mobile app that has also started gathering such data. CalSPEED measures mobile network speeds in California based primarily on drive tests. RootMetrics publishes broadband performance metrics that is largely based on drive test data in 125 U.S. cities and in 50 airports, but also incorporates results of some crowdsourced data. 1. Network Speed a. Ookla 193. Ookla gathers crowdsourced mobile speed data through the use of their Speedtest mobile app. 439 This app is available free of charge to smart phone users, and is designed to test the performance of mobile cellular connections including LTE, 4G, 3G, EDGE, and EVDO networks. Once the app is downloaded, the user can periodically measure the speed of their wireless connection. This data are then used to produce Ookla’s Net Index dataset.440 Because the speed tests rely on the phone’s connection to the server, such factors as congestion, location of the server, proximity and access to a cell tower, and phone quality can affect the result. As presented below and in the Appendices, the Ookla data show significant variation in different geographies, as well as among service providers. Our analysis is based on the speed test source data that Ookla, proprietor of speedtest.net, makes publicly available on its Net Index website.441 194. In this Report, we present the nationwide median download and upload speeds by provider, based on all available U.S. cities in the dataset with sufficient sample size. Chart VI.C.1 presents Ookla’s median download speed measurements for the four nationwide providers from 2012 to June 2014.442 As seen from the chart, upload and download speeds vary by provider.443 AT&T and Verizon Wireless had the two highest download speeds in 2013. Sprint’s users reported the lowest median download speeds during the reporting period. T-Mobile experienced significant speed improvements throughout the reporting period, supplying the second highest median download speeds during the first half of 2014. 438 By contrast, crowdsourced fixed broadband speed data, such as those collected by the FCC through SamKnows, can be gathered with more control. The SamKnows whiteboxes are able to measure actual fixed network speed and are not dependent on the vintage of the client hardware or software. Additionally, the testers are chosen according to a valid sampling technique. 439 http://www.speedtest.net/mobile/. Website accessed 4/18/2014 440 Additional aspects of the Ookla test methodologies are discussed in Appendix VI.C. 441 http://www.netindex.com/. Accessed 4/9/2014. The Ookla dataset used in this report is based on daily mobile download and upload speed data by city and provider for 2013 – June 2014. We drop cities that do not have sufficient observations as well as outliers. Also see Appendix VI.C.i infra 442 More details can be found in Appendix Tables VI.C.i – VI.C.iv 443 One factor that may lead to speed differences between wireless providers is the composition of currently used smartphones. The Ookla Speedtest application is available for download on iOS, Android, or Windows Phones. Each of these operating systems has evolved over time. Vintage smartphones, which do not support 4G, or possibly even 3G service, are still in use. Based on current and past promotions and partnerships, each wireless provider may have a customer base with a different smartphone profile, which can directly affect speed measurements. 97 Federal Communications Commission DA 14-1862 Source: Net Index data by Ookla, 2013 – June 2014 b. FCC 195. In September 2012, the FCC announced that it was expanding its Measuring Broadband America program to include information on mobile broadband service performance in the United States using a crowdsourced approach.444 The program uses the FCC Speed Test app for Android and iPhone devices to test the speed and performance of volunteers' smartphone mobile broadband services.445 The FCC Speed Test app is available free of charge for Android phones and for the iPhone. The FCC speed test can be set to run automatically in the background on Android phones, but iPhone users must execute the speed test manually. This app allows users to measure their mobile broadband performance and voluntarily report these data to the FCC. Collected data include upload and download speed, latency and packet loss, as well as the wireless performance characteristics of the broadband connection and the kind of handsets and versions of operating systems tested. Several other passive metrics are also recorded, including signal strength of the connection, and device manufacturer and model.446 The FCC Speed Test app provides the benefits, and has the limitations, described above for crowdsourced mobile data. 196. The results reported here do not incorporate all attempted tests reported to the Commission. Each individual test report includes information on whether the test completed successfully, whether it timed out due to connection problems, and whether any data from the test are missing. Missing data from a test reflect issues in the operation of the app for that particular test, so that particular test observation is dropped from the dataset. If the test timed out, we did not include speed observations, thereby reporting network speeds only in situations where the test completed successfully. Tests where the user was on a Wifi network were filtered out. We then used a standard econometric test to identify outliers in the data and dropped those outlier observations.447 Based 444 http://www.fcc.gov/measuring-broadband-america/mobile, visited 7/10/2014 445 See http://ldevndj-web01:8080/. The data collected includes speed, latency, and packet loss for both upload and download. . 446 http://www.fcc.gov/measuring-broadband-america/mobile 447 We use the hadimvo test in STATA to identify the outliers. This test identifies multiple outliers in multivariate data using the method of Hadi (1992, 1993), See STATA Manual. 8.99 9.04 1.92 6.16 10.74 9.79 2.90 9.89 0.0 2.0 4.0 6.0 8.0 10.0 12.0 Verizon Wireless AT&T Sprint PCS T-Mobile M ed ia n D ow nl oa d Sp ee d (M bp s) Provider Chart VI.C.1 Median Download Speeds by Provider (Mbps) 2013 - June 2014 2013 2014 (January- June) 98 Federal Communications Commission DA 14-1862 on the remaining data, we calculated the overall mean and median download and upload speeds by service provider.448 197. Based on nationwide FCC Speed Test App data for the time period between November 2013, and July 11, 2014, Verizon Wireless had the highest mean nationwide upload and download speeds, followed by T- Mobile, AT&T, and Sprint. For the median speed, T-Mobile had the highest download speed followed by Verizon and AT&T, while Verizon had the highest median upload speed followed by T-Mobile and AT&T. We see similar patterns in the California data as well. We present the FCC data for the U.S. in Table VI.C.2 and for California in Table VI.C.3.449 Table VI.C.2 FCC - Mean and Median Download and Upload Speeds by Provider, Nationwide November 2013 – July 2014 Service Provider Mean Download Speed Median Download Speed Number of Download Observations Mean Upload Speed Median Upload Speed Number of Upload Observations Verizon Wireless 13.79 9.41 154,106 5.56 3.65 131,836 AT&T 9.83 6.33 99,426 4.20 2.00 83,019 Sprint 3.86 1.40 120,385 1.66 0.78 100,566 T-Mobile 12.76 9.93 142,998 5.05 2.32 121,569 Note: Data from FCC Measuring Mobile Broadband America data. Table based on staff calculations. Under this methodology, only tests that were fully completed (“TRUE” value) were included in the data. All other test results were excluded. Outliers have been dropped. The data were extracted in July 2014. Table VI.C.3 FCC - Mean and Median Download and Upload Speeds by Provider, California November 2013 – July 2014 Service Provider Mean Download Speed Median Download Speed Number of Download Observations Mean Upload Speed Median Upload Speed Number of Upload Observations Verizon Wireless 12.85 7.99 21,414 5.65 3.64 18,724 AT&T 9.14 6.03 14,160 3.94 1.36 12,219 Sprint 3.08 0.98 15,709 1.41 0.74 13,438 T-Mobile 12.20 9.29 23,826 5.81 2.38 20,012 Note: Data from FCC Measuring Mobile Broadband America data. Table based on staff calculations. Under this methodology, only tests that were fully completed (“TRUE” value) were included in the data. All other test results were excluded. Outliers have been dropped. The data were extracted in July 2014. c. RootMetrics 198. RootMetrics runs a test program that measures mobile data, call, and text performance in all 50 states across the United States. Tests are conducted in the 125 most populous metropolitan markets and within the 50 busiest U.S. airports. Each location is tested twice a year by, using the latest Android smartphone available from each provider. All tests, which are conducted solely on the networks of the four nationwide providers, are 448 We describe in more detail the FCC Speed Test methodologies, and present some additional results, in Appendix VI.C. 449 More details can be found in Appendix Tables VI.C.v and VI.C.vi and the associated writeup. 99 Federal Communications Commission DA 14-1862 performed identically across all operators’ devices.450 199. According to RootMetrics, tests are performed during all hours of the day, every day of the week, and nearly every week of the year, but the testing schedule is weighted more heavily toward typical consumer usage hours. Performance is measured indoors and outdoors at the same randomly chosen locations, and drive testing takes place during travel between locations. Results are reported at the national, state, and metro levels. U.S. Census Places are divided into groups by population size, and each population-based group is given equal weighting in the results.451 At the airport level, data collection is conducted within three major publicly accessible areas within each airport: check-in, baggage claim, and various domestic terminals behind security. Efforts are made to avoid well-known busy travel days, such as major holidays.452 200. The RootMetrics Speed Index takes into account speed measurements of both data and texts.453 These results are combined and converted into scores using a proprietary algorithm. RootScores are meant to reflect a consumer’s experience of network performance and are scaled from 0 – 100,454 with the lower limit representing network performance that would result in a poor consumer experience and the upper limit reflecting extraordinary performance.455 The publicly available RootMetrics results are not directly comparable with the Ookla, FCC,,or CalSPEED data, as RootMetrics reports its proprietary speed index as opposed to actual mean and median speeds. However, one can compare the rankings of mobile service providers based on their relative speed performance. Table VI.C.4 provides the national Speed Index data for the second half of 2013 and the first half of 2014.456 Table VI.C.5 provides the same Speed Index data for California457 as a comparison with the CalSPEED speed test data for California. Table VI.C.4 RootMetrics: National Speed Index Data 2nd Half 2013, 1st Half 2014 Service Provider 2nd Half 2013 1st Half 2014 Verizon 88.5 75.7 AT&T 88.7 71.1 Sprint 65.8 54.2 T-Mobile 74.3 64.4 Source: RootMetrics Data, 2013, 2014. Total tests=4,666,641 450 http://www.rootmetrics.com/us/methodology In addition to the performance scores at each location, an Online Coverage Map is available (http://webcoveragemap.rootmetrics.com/us) This map incorporates the sample data described above, along with crowdsourced data that is available through consumer use of the free CoverageMap app, available on Android and iOS. 451 http://www.rootmetrics.com/us/methodology 452 http://www.rootmetrics.com/us/methodology 453 http://www.rootmetrics.com/us/methodology 454 Prior to January 2014, Data RootScores in Metro and Airport RootScore Reports could exceed 100 if performance was extraordinary, http://www.rootmetrics.com/us/methodology 455 http://www.rootmetrics.com/us/standards 456 Source: http://www.rootmetrics.com/us/rsr/united-states/2013/2H and http://www.rootmetrics.com/us/rsr/united- states/2014/1H. For detailed data on the index see Appendix Tables VI.c.viii and VI.c.ix 457 Source: Rootmetrics State Root Score Report found at - http://www.rootmetrics.com/us/rsr/united-states/2013/2H and http://www.rootmetrics.com/us/rsr/california/2014/1H 100 Federal Communications Commission DA 14-1862 Table VI.C.5 Rootmetrics: California Speed Index Data 2nd Half 2013, 1st Half 2014 Service Provider 2nd Half 2013 1st Half 2014 Verizon 84.4 88.9 AT&T 80.3 83.0 Sprint 58.9 62.3 T-Mobile 70.8 83.0 Source: RootMetrics Data, 2013, 2014. Total tests = 436,383. d. CalSPEED 201. CalSPEED is an open source, non-proprietary, network performance measurement tool and methodology created for the California Public Utilities Commission with the assistance of a grant from the National Telecommunications and Information Administration.458 The CalSPEED data presented in this Report is the result of a structured sampling program of 1,986 locations (originally 1,200) scattered throughout California. These sites are visited every six months and tests are run on both the latest Android phones and a USB network device on a Windows based netbook, for each of the four major providers. CalSPEED has now had five rounds of sampling in California.459 202. The CalSPEED data used in this Report were collected from the Spring of 2013 through Spring 2014.460 For our analysis of the data, we dropped any observation that was not in the provider’s coverage area, or any observation that was terminated by the tester. Any other errors are counted as zero throughput. Similar to our analysis of the FCC data, we use an econometric test to identify and drop the outliers in the download and upload speed data from CalSPEED.461 We then calculate the overall mean throughput by provider for three time periods -- the whole of 2013, and Fall 2013 and Spring 2014 combined, and Spring 2014. The second time period corresponds with the FCC speed tests data availability, and makes the two datasets comparable in terms of the time covered. For most of the 2012-2013 period, Verizon Wireless had the fastest mean upload and download speeds, followed by AT&T, T-Mobile, and finally Sprint. In Spring 2014, T-Mobile surpassed AT&T in both mean and median download and upload speeds. Mean download and upload speeds are displayed below in Tables VI.C.6 and VI.C.7, respectively.462 458 More discussion of the CalSPEED dataset is available in Appendix VI.C and Appendix Tables VI.C.vii and VI.C.viii 459 CalSPEED: California Mobile Broadband - An Assessment. Ken Biba Managing Director and CTO Novarum, Inc. 460 Spring 2013 tests were taken between the dates of 4/4/2013 to 4/29/2013, while Fall 2013 tests were taken between the dates of 10/17/2013 to 12/18/2013. Spring 2014 tests were taken between the dates of 4/10/2014 and 6/05/2014. 461 See Section VI.C.1 and footnote 505 infra 462 CalSPEED: California’s Mobile Broadband Assessment, Fall 2014 (Spring 2014 measureent data). Ken Biba, Novarum, Inc. November 2014. Charts VI.C.3 - VI.C.5 taken directly from pages 6-8. 101 Federal Communications Commission DA 14-1862 Table VI.C.6 CalSPEED - Estimated Download Speeds by Provider Service Provider 2013 Fall 2013 and Spring 2014 Spring 2014 Mean Down load Speed (Mbps) Median Down load Speed (Mbps) Number of Tests Mean Down load Speed (Mbps) Median Down load Speed (Mbps) Number of Tests Mean Down load Speed (Mbps) Median Down load Speed (Mbps) Number of Tests Verizon 8.08 6.48 6,127 9.73 8.54 7,506 10.46 9.71 3,697 AT&T 4.99 3.30 5,784 6.21 4.03 7,116 6.65 4.64 3,471 Sprint 0.63 0.40 3,619 2.56 0.75 5,282 2.90 0.84 2,623 T-Mobile 4.35 0.85 4,099 6.02 3.34 5,307 7.16 5.37 2,534 Note: The calculations are based on the CalSPEED data. Outliers have been identified and dropped using an outlier test. 2013 CalSPEED tests include fall and spring tests. Spring 2013 tests were taken between the dates of 4/4/2013 to 4/29/2013, while Fall 2013 tests were taken between the dates of 10/17/2013 to 12/18/2013. Spring 2014 tests were taken between the dates of 4/10/2014 and 6/05/2014. Table VI.C.7 CalSPEED - Estimated Upload Speeds by Provider Service Provider 2013 Fall 2013 and Spring 2014 Spring 2014 Mean Upload Speed (Mbps) Median Upload Speed (Mbps) Number of Tests Mean Upload Speed (Mbps) Median Upload Speed (Mbps) Number of Tests Mean Upload Speed (Mbps) Median Upload Speed (Mbps) Number of Tests Verizon 5.01 3.92 6,132 5.65 5.19 7,508 6.07 5.95 3,697 AT&T 2.34 1.14 5,827 3.09 1.18 7,133 3.39 1.22 3,477 Sprint 0.61 0.63 3,808 1.46 0.79 5,296 1.67 0.81 2,629 T-Mobile 2.16 0.60 4,101 3.77 1.07 5,307 4.82 2.07 2,534 Note: The calculations are based on the CalSPEED data. Outliers have been identified and dropped using an outlier test. 2013 CalSPEED tests include fall and spring tests. Spring 2013 tests were taken between the dates of 4/4/2013 to 4/29/2013, while Fall 2013 tests were taken between the dates of 10/17/2013 to 12/18/2013. Spring 2014 tests were taken between the dates of 4/10/2014 and 6/05/2014. e. Other 203. Speed measurements are also performed by other entities such as by PC Magazine. PC Magazine uses a drive testing methodology. To run the test, they used field-test software from network testing firm “Sensorly,” loaded onto Android-powered smartphones. They tested speeds in 30 different cities during 2013 and 2014 In each city, the drivers stopped in eleven locations for at least 15 minutes (five test cycles) each. They also collected data while moving, both within and between cities. The final speed score is a weighted average balancing 70 percent speed and 30 percent network reliability. In 2013, based on drive tests, PC Magazine ranked AT&T’s LTE network as the fastest overall, followed closely by Verizon LTE. The next in order of speed were T-Mobile HSPA, AT&T HSPA, Sprint LTE, Verizon 3G, and finally Sprint 3G. 463 In 2014, Nielsen ranked Verizon LTE as the highest with an average download speed of 19.6 Mbps. T-Mobile followed with 16.8 Mbps and AT&T had 11.9 Mbps, and Sprint had 4.4 Mbps.464 Nielsen uses a crowdsourced application similar to Ookla and the FCC to measure mobile provider speed. 463 See http://www.pcmag.com/article2/0,2817,2420334,00.asp, visited 7/17/2014 464 Nielsen testmy.net speed test data. Visited on Nov. 6, See 2014http://www.pcmag.com/article2/0,2817,2459186,00.asp 102 Federal Communications Commission DA 14-1862 2. Latency 204. Latency refers to several types of delays typically incurred during network data processing, and is typically measured in milliseconds (ms). One common measure is round-trip latency, which measures the amount of time it takes a data packet to travel from a source to a destination and back.465 Latency is often affected by factors such as the specifics of the cellular network architecture or processing delays that may occur when the packets need to pass through proxy servers.466 205. Ookla Data. Ookla speedtest automatically selects the server with the fastest latency, chosen from an initial set of possible test servers, which may bias subsequent tests towards higher performance. For the given time period (2013 – June 2014), Verizon has the lowest latency, closely followed by AT&T. 206. FCC Data. Similar to Ookla, the FCC test selects the server with the fastest latency, chosen from an initial set of possible test servers, which may bias subsequent tests towards higher performance. It is possible that consecutive tests in the same place, on the same provider, and at about the same time may test to different servers. If a packet is not received back within three seconds of sending, it is treated as lost.467 Based on FCC Speed Test App data as of July , 2014, T-Mobile had the lowest reported latency, at 98 ms. This was followed by Verizon at 102 ms, AT&T at 124 ms, and finally Sprint, at 135 ms. 207. CalSPEED Data. CalSPEED tests the complete network path, from the client device, through the local access network, through the Internet backbone, to two ultimate server destinations. One server is physically located in Northern California and the other in Northern Virginia.468 Based on the CalSPEED data, latency continues to improve, with Verizon and AT&T having the lowest latency, followed by Sprint and then T- Mobile.469 D. Differentiation in Mobile Wireless Handsets/ Devices 208. In addition to competing on price and network quality, mobile wireless providers continue to compete by offering consumers a variety of different mobile wireless devices with innovative features.470 In particular, providers offer a range of data-centric smartphones471 and tablets which are made by different 465 More precisely, it is measured as the sum of time from the start of packet transmission by a source to the start of packet reception by a destination plus the time that it takes for the packet to travel back from the receiving destination to the source. This excludes the amount of time that a destination system spends processing the packet. 466 International Broadband Data Report; http://www.fcc.gov/document/international-broadband-data-report. DA 12-1334 at 14. 467 http://www.fcc.gov/measuring-broadband-america/mobile/technical-summary 468 CalSPEED: California Mobile Broadband - An Assessment. Ken Biba Managing Director and CTO Novarum, Inc. 469 CalSPEED: California’s Mobile Broadband Assessment, Fall 2014 (Spring 2014 measureent data). Ken Biba, Novarum, Inc. November 2014. 470 AT&T Comments at 4, 24, 27; Verizon Comments at 77 471 While there is no industry standard definition of a smartphone, for purposes of this Report we continue to consider the distinguishing features of a smartphone to be: an HTML browser that allows easy access to the full, open Internet; an operating system that provides a standardized interface and platform for application developers; and a larger screen size than a traditional, voice-centric handset. Many smartphones also have touch screens and/or a QWERTY keypad, and run an operating system that offers a standard platform for application developers to create and sell device software through an application store. See Sixteenth Report 28 FCC Rcd at 3821 ¶ 220. By contrast, the basic handset category includes voice- centric handsets that do not allow or are not designed for easy web browsing. In addition to smartphones and basic handsets, a third category of devices consists of data-centric devices that have no inherent voice capability, such as USB wireless modem laptop cards, mobile Wi-Fi devices, e-readers, and laptops and netbooks with embedded mobile wireless modems. 103 Federal Communications Commission DA 14-1862 manufacturers and run different operating systems. Smaller carriers, however, have cited limited access to highly sought-after devices as an impediment to their ability to compete. 472 209. Since Apple entered the smartphone business with the iPhone in June 2007, many handset manufacturers have introduced competing products with similar features such as touch screens, mobile web browsing capabilities, and current-generation operating systems. Popular smartphone operating systems such as the Android and the Apple iOS were available from multiple service providers, permitting consumers to pair their preferred operating systems with various service providers. 210. Although many devices are offered by multiple providers, some providers may offer certain devices only on postpaid plans. For example, from September 2013 to early December 2013, Walmart only sold the iPhone 5s and the iPhone 5c with a two-year contract.473 As of December 13, 2013, Walmart began offering these devices with the purchase of contract-free Straight Talk Wireless and Net10 Wireless network plans (both owned by TracFone).474 For T-Mobile, until recently, the phone selection for prepaid plans was not as extensive or updated as the phone selection for postpaid plans, and the iPhone was excluded from its prepaid selection. In June 2014, T-Mobile announced that it would sell the latest iPhone on its prepaid plans as well, further blurring the lines between postpaid and prepaid price plans. Additionally, for a limited time, T-Mobile also offered $50 cash/gift card incentive to customers who purchased a new iPhone with T-Mobile’s prepaid service. Sprint sells the iPhone for use on its Boost Mobile and Virgin Mobile prepaid brands. AT&T’s Cricket/Aio brand offers a slightly higher-end handset selection, including the Samsung Galaxy S4, the Apple iPhone 5c, and the Apple iPhone 5s 16GB. Cricket plans to offer a wider variety of smartphones, starting at around $50 but also including high-end smartphones including Apple's iPhone 5c and 5s. AT&T’s GoPhone brand however, offers feature phones and lower-end smartphones, such as the iPhone 4, Nokia Lumia 520, and Samsung Galaxy Express. Verizon offers both smartphone and feature phone packages for its monthly prepaid plans. Although the most recently released smartphones are not available with these plans, the iPhone 4 and the iPhone 4s are available for purchase. Tablet and data-only devices are not commonly offered on the prepaid platform. 211. In addition to offering a variety of smartphones and traditional handsets, mobile wireless providers also sell or provide connectivity for other data-only devices such as tablets, e-readers, wireless data cards, mobile Wi-Fi hotspots,475 and netbook computers with embedded modems. The use of data-only devices with mobile network connectivity has grown in recent years.476 Providers compete with one another by offering such devices; and also on the speed, coverage, and price of the mobile data connections on which these devices rely. Mobile wireless providers offer wireless data cards and mobile Wi-Fi hotspots to consumers seeking mobile Internet connections for laptop computers and other Wi-Fi enabled devices. Because such devices tend to have similar characteristics and functionality across equipment manufacturers, providers generally differentiate their offerings of these products based on the speed and coverage of the mobile data networks to which such devices connect, rather than the uniqueness of the devices themselves. 472 See CCA Reply Comments at 8. 473 iPhone 5S and 5C come to Walmart’s no-contract plans, CNet.com, Dec. 9, 2013. http://www.cnet.com/news/iphone-5s- and-5c-come-to-walmarts-no-contract-plans/ (visited July 1, 2014). 474 Id. 475 Mobile Wi-Fi, or “Mi-Fi,” devices are credit card-sized, mobile Wi-Fi routers with mobile broadband wide-area connections that allow a certain number of Wi-Fi-enabled devices in short range to connect to the Internet via a Wi-Fi connection. Many smartphones are now sold with built-in Wi-Fi hotspot capabilities, allowing them to serve as mobile Wi-Fi hotspots for an additional charge. 476 US Wi-Fi Households to Own Average of 11 Wi-Fi Devices in 2017 says Strategy Analytics, Press Release, Strategy Analytics, Feb. 27, 2014. http://www.strategyanalytics.com/default.aspx?mod=pressreleaseviewer&a0=5483 (visited June 17, 2014). 104 Federal Communications Commission DA 14-1862 E. Advertising and Marketing 212. Mobile wireless providers also compete for customers through advertising and marketing, including by establishing retail and distribution networks that reach their target audience. Several providers stated that the goal of their advertising and marketing efforts is to increase and maintain brand awareness and to support distribution.477 Providers may also engage in advertising and marketing either to inform consumers about available products or services or to try to increase sales by influencing consumer preferences.478 Providers may advertise in the media, in internet and mobile applications, in sponsorships and co-branding, and at events.479 213. Overall, the Communications/Telecommunications segment had an annual growth in TV spot advertising of 1.4 percent from 2012 to 2013.480 Mobile advertising is expected to show the largest increase, with advertisers predicted to spend 83 percent more on tablets and smartphones in 2014 than they did in 2013.481 By the end of 2014, mobile is expected to represent almost 10 percent of all media ad spending, surpassing newspapers, magazines, and radio.482 In 2013 and 2014, mobile wireless service providers were quite aggressive with their advertising campaigns.483 Some providers’ marketing campaigns continued to focus on the quality and size of their mobile broadband networks.484 Many providers sought to highlight their network speed, coverage and the data capabilities of devices available on these networks. Some providers promoted the advantages of their particular service plans relative to those of rivals. In Kantar Media’s 2013 rankings of advertising spending, AT&T and Verizon Communications were the third and eighth largest U.S. advertisers, respectively.485 214. In 2013, T-Mobile branded itself as the “Uncarrier” and vowed to “shake-up the industry,” using the advertising slogan: “Don’t play by the rules. Break them. Unleash.”486 It advertised the elimination of contracts and payment of early termination fees for customers who switched from another provider to T-Mobile. T-Mobile continued this plan throughout 2013 and into the beginning of 2014 with “unrelenting attacks on its top 477 See 2013 SEC Form 10-K for Sprint Nextel at 5, and 2013 SEC Form 10-K for US Cellular at 5. 478 See Kyle Bagwell, “The Economic Analysis of Advertising,” in Handbook of Industrial Organization, Volume 3, ed. M. Armstrong and R. Porter, (Elsevier B.V., 2007), at 1705-1706. Mobile wireless service is an “experience good” – a product or service that the customer must consume before determining its quality. See Lynne Pepall, Dan Richards, and George Norman. Industrial Organization (4th ed.), Blackwell Publishing, 2008, at 524. Consequently, information contained in wireless advertising tends to be indirect information. By advertising, a firm may signal that it is efficient, implying that it offers good deals. Advertising may also remind repeat consumers of the quality of an experience good. Finally, since a firm has an incentive to direct its advertising toward the consumers who may value its product the most, a seemingly uninformative advertisement can better match products with buyers. See Kyle Bagwell, “The Economic Analysis of Advertising,” in Handbook of Industrial Organization, Volume 3, ed. M. Armstrong and R. Porter, (Elsevier B.V., 2007), at 1718-1720, 1774-1791; Phillip Nelson, “Advertising as Information,” in Journal of Political Economy, v. 82 (1974) at 729- 754. 479 See, for example, 2013 SEC Form 10-K for Sprint Nextel, and US Cellular. 480 See Historical Analysis of Broadcast Spot Advertising Revenue Categories, SNL Kagan Broadcast Investor, June 19, 2014. 481 See Total US Ad Spending to See Largest Increase Since 2004, eMarketer, July 2, 2014. 482 Id. 483 AT&T Takes on T-Mobile, Verizon with New Marketing Slogan, FierceWireless, Jan. 6, 2014, http://www.fiercewireless.com/story/att-takes-t-mobile-verizon-new-marketing-slogan/2014-01-06 (visited May 28, 2014). 484 See Fourteenth Report, 25 FCC at 11493 ¶ 132. 485 Kantar Media Reports U.S. Advertising Expenditures Increased 0.9 Percent in 2013, Fueled by Larger Advertisers, Kantar Media, March 25, 2014, http://kantarmedia.us/press/kantar-media-reports-us-advertising-expenditures-increased-09- percent-2013, (visited May 28, 2014). 486 See http://www.t-mobile.com/landing/whyt-mobile.html (visited April 29, 2014). 105 Federal Communications Commission DA 14-1862 competitors.”487 For example, in 2014, T-Mobile targeted Verizon Wireless by arguing that Verizon Wireless’ ads on network coverage are misleading.488 T-Mobile’s ads say “Fold up your old map, Verizon,” and claim that T-Mobile covers 96 percent of all Americans with voice and data.489 In the beginning of 2014, AT&T announced a marketing slogan targeting T-Mobile and Verizon Wireless. AT&T tried to differentiate itself from its competitors by advertising, “Bigger than T-Mobile, faster than Verizon Wireless and more reliable than everybody.”490 215. Some providers’ advertisements highlighted network improvements. In 2014, AT&T launched an advertising campaign called “Better Network” to show how it is improving coverage and capacity on its wireless network.491 Likewise, in 2014, Verizon Wireless introduced an advertising campaign named XLTE to make customers aware of improvements in portions of its network.492 In addition to marketing with traditional media, service providers have also advertised their products on the internet, social media, mobile applications, and through sponsorships.493 VII. CONCLUSION 216. Promoting competition is a fundamental goal of the Commission’s policymaking. Competition has played and must continue to play an essential role in the mobile wireless industry – leading to lower prices and higher quality for American consumers, and producing innovation and investment in wireless networks, devices, and services. This Report analyzes competition in the mobile wireless industry pursuant to section 332(c)(1)(C) of the Communications Act and highlights several key trends in the industry. As with past reports, this Report examines various facets of the mobile wireless industry including market concentration, the conduct and rivalry of service providers, industry performance and outcomes, and consumer responses to mobile wireless service offerings. It also analyzes competition in other segments of the mobile wireless ecosystem, including spectrum, backhaul facilities, and handsets/devices and mobile applications. 487 “AT&T Goes After T-Mobile Attack Ads,” March 1, 2013. http://www.bloomberg.com/news/2013-02-28/at-t-goes-after- former-merger-partner-t-mobile-with-attack-ads.html 488 “T-Mobile Takes on Verizon in LTE Advertising and Network Battle,” FierceWireless, March 14, 2014, http://www.fiercewireless.com/story/t-mobile-takes-verizon-lte-advertising-and-network-battle/2014-03-14 (visited May 28, 2014). 489 Id. 490 AT&T Takes on T-Mobile, Verizon with New Marketing Slogan, FierceWireless, Jan. 6, 2014, http://www.fiercewireless.com/story/att-takes-t-mobile-verizon-new-marketing-slogan/2014-01-06 (visited May 28, 2014). 491 AT&T Launches New Ad Campaign “Better Network’ Highlighting Small Cells, DAS, FierceWireless.com, March 17, 2014, http://www.fiercewireless.com/story/att-launches-new-ad-campaign-better-network-highlighting-small-cells-das/2014- 03-17 (visited May 23,2014). 492 Verizon Wireless Names Faster Part of Its Network: XLTE, New York Times, May 19, 2014, http://bits.blogs.nytimes.com/2014/05/19/verizon-wireless-names-faster-part-of-its-network- xlte/?_php=true&_type=blogs&_php=true&_type=blogs&_r=1& (visited May 23, 2014). 493 See Sprint 2013 SEC Form 10-K, at 7. 106 Federal Communications Commission DA 14-1862 VIII. PROCEDURAL MATTERS 217. This Seventeenth Report is issued pursuant to authority contained in Section 332(c)(1)(C) of the Communications Act of 1934, as amended, 47 U.S.C. § 332(c)(1)(C), and authority delegated to the Wireless Telecommunications Bureau under section 0.331 of the Commission’s rules, 47 C.F.R. § 0.331. 218. It is ORDERED that copies of this Report be sent to the appropriate committees and subcommittees of the United States House of Representatives and the United States Senate. FEDERAL COMMUNICATIONS COMMISSION Roger C. Sherman Chief, Wireless Telecommunications Bureau 107 Federal Communications Commission DA 14-1862 APPENDIX I Map Appendix 108 Federal Communications Commission DA 14-1862 APPENDIX II COMPETITIVE DYNAMICS WITHIN THE INDUSTRY Connections, Net Adds and Churn Table II.B.i Estimated Total Mobile Wireless Connections NRUF CTIA Year Connected Devices (millions) Increase from previous year (millions) Connections Per 100 People Estimated Connections (millions) 2001 128.5 N/A 45 128.4 2002 141.8 13.3 49 140.8 2003 160.6 18.8 54 158.7 2004 184.7 24.1 62 182.1 2005 213.0 28.3 71 207.9 2006 241.8 28.8 80 233.0 2007 263.0 21.2 86 255.4 2008 279.6 16.6 91 270.3 2009 290.7 11.1 94 285.6 2010 301.8 11.1 97 296.3 2011 317.3 15.5 102 316.0 2012 329.2 11.9 106 326.5 2013 339.2 10.0 109.9 335.7 Note: Based on CTIA Year-End 2013 Wireless Indices Report, table 6. NRUF 2001 – 2013. 2010 Census data (Nationwide Population = 308,745,538) Table II.B.ii Quarterly Total Mobile Wireless Connections by Service Segment 2011 – 2nd Quarter 2014 Quarter Year Postpaid Prepaid Wholesale Connected Devices Total Connections 1Q11 213,375 62,412 9,017 21,035 305,838 2Q11 213,967 63,506 10,034 22,487 309,995 3Q11 214,412 65,238 11,153 23,933 314,736 4Q11 215,827 67,236 12,659 24,009 319,731 1Q12 215,466 69,133 13,955 24,502 323,056 2Q12 215,633 70,649 13,423 24,982 324,687 3Q12 216,129 71,112 13,567 25,836 326,644 4Q12 218,246 71,728 13,416 26,889 330,279 1Q13 217,887 73,007 16,847 28,233 335,974 2Q13 218,473 71,687 17,445 29,551 337,156 3Q13 221,142 71,906 17,881 30,932 341,862 4Q13 223,759 72,978 18,683 31,958 347,378 1Q14 225,580 74,827 17,738 33,661 351,807 2Q14 228,348 73,875 18,713 35,234 356,170 Source: UBS Investment Research. USB Wireless 411 Version 51, Figure 17: UBS 411 Version 54. 109 Federal Communications Commission DA 14-1862 Table II.B.iii Quarterly Net Adds in Mobile Wireless Connections by Service Segment (In thousands) 2009 – 1st Half 2014 Quarter Year Postpaid Prepaid Wholesale Connected Devices Total Net Adds 1Q09 684 2,675 308 244 3,911 2Q09 689 1,530 (50) 758 2,927 3Q09 1,173 1,018 132 184 2,507 4Q09 1,175 2,792 306 1,345 5,618 1Q10 (6) 2,464 671 1,237 4,366 2Q10 813 749 483 1,421 3,466 3Q10 823 1,565 607 1,634 4,629 4Q10 895 2,633 39 1,831 5,398 1Q11 196 2,661 1,210 1,725 5,791 2Q11 787 1,093 1,017 1,452 4,349 3Q11 583 1,730 1,119 1,446 4,878 4Q11 1,304 1,998 1,506 76 4,884 1Q12 (147) 1,891 1,296 493 3,533 2Q12 784 414 568 480 2,246 3Q12 (405) 462 1,244 854 2,156 4Q12 2,177 603 (151) 1,053 3,682 1Q13 (3,872) 1,278 3,431 1,344 2,181 2Q13 263 (1,391) 598 1,318 788 3Q13 1,132 280 436 1,381 3,229 4Q13 2,492 1,069 802 1,026 5,389 1Q14 3,147 472 -945 1,703 4,378 2Q14 2,899 (1,029) 975 1,574 4,418 Source: 1Q09 - 4Q10 data from Sixteenth Competition Report 28 FCC Rcd at 3836, UBS Wireless 411 4Q11 at 10. 1Q12 - 2Q14 data from UBS Wireless 411: Version 54. UBS categorizes Tracfone customers as prepaid, not wholesale. 110 Federal Communications Commission DA 14-1862 Market Share and Concentration Table II.C.i Market Concentration by EA, 2011 – 2013 HHI 2013 Rank EA Market Name 2011 2012 2013 1 142 Scottsbluff, NE-WY 6618 6606 6689 2 144 Billings, MT-WY 6305 6417 6557 3 145 Great Falls, MT 6315 6571 6414 4 116 Sioux Falls, SD-IA-MN-NE 6028 6150 6297 5 111 Minot, ND 5542 5933 6033 6 110 Grand Forks, ND-MN 5549 5672 5948 7 115 Rapid City, SD-MT-ND-NE 5570 5643 5894 8 113 Fargo-Moorhead, ND-MN 5354 5458 5624 9 143 Casper, WY-ID-UT 5745 5708 5611 10 119 Lincoln, NE 5015 5076 5348 11 14 Salisbury, MD-DE-VA 5152 5122 5293 12 117 Sioux City, IA-NE-SD 4532 4589 5051 13 56 Toledo, OH 4847 4770 5005 14 149 Twin Falls, ID 4504 4444 4997 15 85 Lafayette, LA 4512 4795 4915 16 95 Jonesboro, AR-MO 4376 4583 4909 17 36 Dothan, AL-FL-GA 4128 4508 4861 18 45 Johnson City-Kingsport-Bristol, TN-VA 4182 4403 4807 19 148 Idaho Falls, ID-WY 4534 4432 4782 20 42 Asheville, NC 4357 4497 4763 21 4 Burlington, VT-NY 4704 4658 4700 22 38 Macon, GA 4358 4136 4661 23 72 Paducah, KY-IL 5116 4931 4616 24 109 Duluth-Superior, MN-WI 4464 4587 4567 25 37 Albany, GA 3801 3821 4522 26 7 Rochester, NY-PA 4343 4373 4508 27 154 Flagstaff, AZ-UT 4287 4203 4434 28 139 Santa Fe, NM 4544 4520 4409 29 92 Fayetteville-Springdale-Rogers, AR-MO-OK 4428 4306 4407 30 55 Cleveland-Akron, OH-PA 3948 4104 4404 31 54 Erie, PA 4159 4161 4352 32 52 Wheeling, WV-OH 4169 4203 4335 33 105 La Crosse, WI-MN 4031 4131 4310 111 Federal Communications Commission DA 14-1862 34 90 Little Rock-North Little Rock, AR 3526 3823 4241 35 69 Evansville-Henderson, IN-KY-IL 4190 4061 4237 36 91 Fort Smith, AR-OK 3628 3842 4201 37 9 State College, PA 4648 4132 4187 38 6 Syracuse, NY-PA 3989 4044 4185 39 120 Grand Island, NE 6397 5707 4176 40 84 Baton Rouge, LA-MS 4114 4232 4166 41 39 Columbus, GA-AL 3553 3643 4145 42 43 Chattanooga, TN-GA 3755 3851 4124 43 75 Tupelo, MS-AL-TN 4375 4346 4097 44 106 Rochester, MN-IA-WI 3770 3882 4085 45 86 Lake Charles, LA 3487 3728 4070 46 35 Tallahassee, FL-GA 3540 3657 4046 47 89 Monroe, LA 3482 3723 3973 48 118 Omaha, NE-IA-MO 3558 3641 3969 49 98 Columbia, MO 3918 3912 3955 50 1 Bangor, ME 3783 3852 3944 51 51 Columbus, OH 3417 3520 3893 52 171 Anchorage, AK 3971 4083 3863 53 97 Springfield, IL-MO 3713 3914 3851 54 147 Spokane, WA-ID 3586 3620 3841 55 68 Champaign-Urbana, IL 3682 3710 3839 56 162 Fresno, CA 2953 2989 3787 57 27 Augusta-Aiken, GA-SC 3426 3410 3762 58 48 Charleston, WV-KY-OH 3618 3626 3749 59 74 Huntsville, AL-TN 3329 3441 3743 60 41 Greenville-Spartanburg-Anderson, SC-NC 3525 3426 3739 61 79 Montgomery, AL 3325 3496 3725 62 101 Peoria-Pekin, IL 3474 3552 3697 63 87 Beaumont-Port Arthur, TX 3383 3469 3692 64 28 Savannah, GA-SC 2766 2925 3671 65 165 Redding, CA-OR 3299 3405 3621 66 5 Albany-Schenectady-Troy, NY 3524 3518 3617 67 88 Shreveport-Bossier City, LA-AR 3263 3456 3612 68 66 Fort Wayne, IN 3551 3471 3601 69 53 Pittsburgh, PA-WV 3256 3310 3587 70 8 Buffalo-Niagara Falls, NY-PA 3255 3303 3586 71 94 Springfield, MO 3641 3600 3565 72 135 Odessa-Midland, TX 3474 3526 3560 73 168 Pendleton, OR-WA 3337 3397 3542 112 Federal Communications Commission DA 14-1862 74 150 Boise City, ID-OR 3226 3273 3527 75 128 Abilene, TX 3687 3635 3499 76 151 Reno, NV-CA 3110 3175 3474 77 24 Columbia, SC 3260 3166 3471 78 15 Richmond-Petersburg, VA 3338 3362 3467 79 47 Lexington, KY-TN-VA-WV 3548 3342 3467 80 46 Hickory-Morganton, NC-TN 2973 3001 3447 81 23 Charlotte-Gastonia-Rock Hill, NC-SC 3068 3120 3440 82 140 Pueblo, CO-NM 3418 3317 3436 83 26 Charleston-North Charleston, SC 3076 3152 3430 84 132 Corpus Christi, TX 2306 2411 3412 85 11 Harrisburg-Lebanon-Carlisle, PA 3445 3406 3362 86 125 Oklahoma City, OK 2988 2825 3316 87 25 Wilmington, NC-SC 2940 3044 3296 88 50 Dayton-Springfield, OH 2722 2774 3291 89 65 Elkhart-Goshen, IN-MI 3170 3158 3287 90 134 San Antonio, TX 2403 2444 3276 91 2 Portland, ME 3036 3057 3260 92 100 Des Moines, IA-IL-MO 3046 3081 3252 93 67 Indianapolis, IN-IL 3199 3163 3247 94 22 Fayetteville, NC 2885 2947 3245 95 155 Farmington, NM-CO 3276 3191 3232 96 104 Madison, WI-IA-IL 3276 3281 3224 97 62 Grand Rapids-Muskegon-Holland, MI 3088 3142 3217 98 156 Albuquerque, NM-AZ 3006 3032 3216 99 57 Detroit-Ann Arbor-Flint, MI 2800 2902 3193 100 124 Tulsa, OK-KS 2926 2635 3184 101 19 Raleigh-Durham-Chapel Hill, NC 2854 2949 3174 102 44 Knoxville, TN 2714 2815 3135 103 126 Western Oklahoma, OK 2512 2861 3132 104 78 Birmingham, AL 3035 3042 3129 105 81 Pensacola, FL 2979 3054 3109 106 130 Austin-San Marcos, TX 2723 2754 3100 107 96 St. Louis, MO-IL 2728 2782 3083 108 93 Joplin, MO-KS-OK 3270 3252 3083 109 18 Greensboro-Winston-Salem-High Point, NC-VA 2770 2781 3080 110 158 Phoenix-Mesa, AZ-NM 2778 2813 3080 111 76 Greenville, MS 2958 3150 3075 112 159 Tucson, AZ 2792 2779 3070 113 80 Mobile, AL 3150 3149 3067 113 Federal Communications Commission DA 14-1862 114 20 Norfolk-Virginia Beach-Newport News, VA-NC 2823 2882 3063 115 167 Portland-Salem, OR-WA 2711 2771 3036 116 71 Nashville, TN-KY 2584 2645 3016 117 77 Jackson, MS-AL-LA 3031 3079 3016 118 73 Memphis, TN-AR-MS-KY 2507 2585 3010 119 102 Davenport-Moline-Rock Island, IA-IL 2779 2819 3000 120 83 New Orleans, LA-MS 3189 3175 2976 121 3 Boston-Worcester-Lawrence-Lowewell-Brockton, MA-NH 2841 2843 2967 122 123 Topeka, KS 2843 2882 2967 123 136 Hobbs, NM-TX 2966 3207 2957 124 161 San Diego, CA 2581 2637 2913 125 163 San Francisco-Oakland-San Jose, CA 2720 2742 2899 126 133 McAllen-Edinburg-Mission, TX 2823 2533 2899 127 107 Minneapolis-St. Paul, MN-WI-IA 2815 2832 2898 128 13 Washington-Baltimore, DC-MD-VA-WV-PA 2695 2735 2891 129 166 Eugene-Springfield, OR-CA 2651 2704 2886 130 164 Sacramento-Yolo, CA 2727 2741 2882 131 49 Cincinnati-Hamilton, OH-KY-IN 2437 2543 2870 132 141 Denver-Boulder-Greeley, CO-KS-NE 2479 2533 2869 133 122 Wichita, KS-OK 2821 2826 2868 134 70 Louisville, KY-IN 2545 2549 2846 135 31 Miami-Fort Lauderdale, FL 2286 2292 2839 136 21 Greenville, NC 2602 2669 2833 137 12 Philadelphia-Wilmington-Atlantic City, PA-NJ-DE-MD 2619 2612 2831 138 127 Dallas-Fort Worth, TX-AR-OK 2617 2681 2829 139 103 Cedar Rapids, IA 2645 2666 2817 140 170 Seattle-Tacoma-Bremerton, WA 2763 2778 2816 141 40 Atlanta, GA-AL-NC 2570 2602 2778 142 33 Sarasota-Bradenton, FL 2640 2639 2771 143 16 Staunton, VA-WV 2740 2726 2755 144 169 Richland-Kennewick-Pasco, WA 2803 2761 2729 145 157 El Paso, TX-NM 2309 2349 2717 146 10 New York-North New Jersey-Long Island, NY-NJ-CT-PA 2582 2591 2702 147 29 Jacksonville, FL-GA 2440 2461 2695 148 131 Houston-Galveston-Brazoria, TX 2318 2357 2687 149 152 Salt Lake City-Ogden, UT-ID 2460 2476 2685 150 138 Amarillo, TX-NM 2857 2774 2685 151 160 Los Angeles-Riverside-Orange County, CA-AZ 2415 2437 2634 152 34 Tampa-St. Petersburg-Clearwater, FL 2247 2262 2632 153 137 Lubbock, TX 2792 2704 2629 114 Federal Communications Commission DA 14-1862 154 60 Appleton-Oshkosh-Neenah, WI 2510 2488 2602 155 32 Fort Myers-Cape Coral, FL 2402 2433 2595 156 30 Orlando, FL 2392 2401 2578 157 172 Honolulu, HI 2497 2528 2559 158 153 Las Vegas, NV-AZ-UT 2139 2211 2557 159 99 Kansas City, MO-KS 2357 2388 2553 160 59 Green Bay, WI-MI 2478 2515 2545 161 17 Roanoke, VA-NC-WV 2375 2387 2510 162 108 Wausau, WI 2010 2008 2474 163 82 Biloxi-Gulfport-Pascagoula, MS 2491 2501 2473 164 64 Chicago-Gary-Kenosha, IL-IN-WI 2120 2180 2360 165 129 San Angelo, TX 2056 2078 2317 166 63 Milwaukee-Racine, WI 2119 2143 2237 112 Bismarck, ND-MT-SD 5809 5963 * 58 Northern Michigan, MI * * * 61 Traverse City, MI * * * 114 Aberdeen, SD * * * 121 North Platte, NE-CO * * * 146 Missoula, MT 7178 * * Note: Based on NRUF and 2010 census data, EAs defined as in 1995. 115 Federal Communications Commission DA 14-1862 Financial Indicators Table II.D.i Annualized Average Revenue Per Reported Subscriber Unit (ARPU) 1993 – 2013 Year Total Annual Service Revenue Percentage Change Average Reported Subscribers Average Monthly Revenue per Active Subscriber Unit 1993 $10,895,174,566 11,861,362 $76.55 1994 $14,229,921,264 30.6% 18,299,487 $64.80 1995 $19,081,239,000 34.1% 26,757,320 $59.43 1996 $23,634,971,000 23.9% 35,554,818 $55.40 1997 $27,485,632,936 16.3% 46,375,849 $49.39 1998 $33,133,174,978 20.6% 58,455,471 $47.23 1999 $40,018,489,104 20.8% 71,885,076 $46.39 2000 $52,466,019,720 31.1% 90,048,320 $48.55 2001 $65,316,235,000 24.5% 109,318,848 $49.79 2002 $76,508,187,000 17.1% 125,002,023 $51.00 2003 $87,624,093,000 14.5% 141,658,059 $51.55 2004 $102,121,210,043 16.5% 161,980,026 $52.54 2005 $113,538,220,438 11.2% 186,801,940 $50.65 2006 $125,456,824,884 10.5% 213,077,033 $49.07 2007 $138,869,303,958 10.7% 234,921,960 $49.26 2008 $148,084,169,893 6.6% 252,539,475 $48.87 2009 $152,551,853,953 3.0% 265,038,212 $47.97 2010 $159,929,646,977 4.9% 280,392,201 $47.53 2011 $169,767,314,353 6.2% 306,840,648 $46.11 2012 $185,013,934,995 9.0% 314,685,754 $48.99 2013 $189,192,811,836 2.3% 323,133,932 $48.79 Note: Based on CTIA Year-End 2013 Wireless Indices Report, Table 27. 116 Federal Communications Commission DA 14-1862 Table II.D.ii Change in CPI, 1997 - 2013 Year CPI Wireless Telephone Services CPI Telephone Services CPI Land-line Telephone Services CPI Index Value Annual Change Index Value Annual Change Index Value Annual Change Index Value Annual Change 1997 100 100 100 - - 1998 101.6 95.1 100.7 - - 1999 103.8 2.2% 84.9 -10.7% 100.1 -0.6% - - 2000 107.3 3.4% 76.0 -10.5% 98.5 -1.6% - - 2001 110.3 2.8% 68.1 -10.4% 99.3 0.8% - - 2002 112.1 1.6% 67.4 -1.0% 99.7 0.4% - - 2003 114.6 2.3% 66.8 -0.9% 98.3 -1.4% - - 2004 117.7 2.7% 66.2 -0.9% 95.8 -2.5% - - 2005 121.7 3.4% 65.0 -1.8% 94.9 -0.9% - - 2006 125.6 3.2% 64.6 -0.6% 95.8 0.9% - - 2007 129.2 2.8% 64.4 -0.3% 98.2 2.6% - - 2008 134.1 3.8% 64.2 -0.2% 100.5 2.2% - - 2009 133.7 -0.4% 64.3 0.0% 102.4 1.9% - - 2010 135.9 1.6% 62.4 -2.9% 102.4 0.0% 101.6 - 2011 140.1 3.2% 60.1 -3.6% 101.2 -1.1% 103.3 1.7% 2012 143.0 2.1% 59.4 -1.2% 101.7 .04% 106.1 2.7% 2013 145.2 1.5% 58.2 -1.6% 101.6 0% 109.3 3.0% 1997 to 2013 34.0% -42.7% 1.2% 7.4% Note: Data from Bureau of Labor Statistics. All CPI figures were taken from BLS databases found on the BLS Internet site available at http://www.bls.gov. Beginning in January 2010, the CPIs for local telephone service and long-distance telephone service were discontinued and replaced by a new CPI for land-line telephone services. 117 Federal Communications Commission DA 14-1862 APPENDIX III OVERALL WIRELESS INDUSTRY METRICS Network Coverage The tables below are based on Commission estimates derived from census block analysis of Mosaik CoverageRight coverage maps, January 2014. Population data are from the 2010 Census, and include the United States and Puerto Rico. Square miles include the United States and Puerto Rico. There are approximately 11 million census blocks and 312 million people in the entire United States (based on the 2010 Census). We note that the percentages of population located in census blocks where zero, one, two, or three or more mobile broadband providers represent network coverage, which does not necessarily mean that they offered service to residents in the census block. In addition, we emphasize that a provider reporting mobile broadband coverage in a particular census block may not provide coverage everywhere in the census block. For both these reasons, the number of providers in a census block does not necessarily reflect the number of choices available to a particular individual or household, and does not purport to measure competition. In addition, calculations based on Mosaik data on coverage, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless and mobile broadband coverage. Table III.A.i Estimated Mobile Wireless Coverage by Census Block Including Federal Land, Jan. 2014 Number of Providers with Coverage in a Block Number of Blocks (Thousands) POPs Contained in Those Blocks % of Total US POPs Square Miles Contained in Those Blocks % of Total US Square Miles Road Miles Contained in Those Blocks % of Total US Road Miles US Total 11,155,486 312,471,327 100.0 3,802,067 100.0 6,821,187 100.0 1 or more 10,941,378 312,063,148 99.9 2,846,332 74.9 6,516,291 95.5 2 or more 10,607,309 310,530,748 99.4 2,466,422 64.9 6,063,229 88.9 3 or more 9,573,697 302,526,668 96.8 1,790,407 47.1 4,971,642 72.9 4 or more 8,075,773 285,562,448 91.4 1,134,924 29.9 3,692,196 54.1 5 or more 2,586,130 71,229,295 22.8 394,447 10.4 1,251,742 18.4 Note: Based on January 2014 Mosaik Data and 2010 Census Data. The number of providers in a census block represent network coverage, which does not necessarily reflect the number of choices available to a particular individual or household, Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. 118 Federal Communications Commission DA 14-1862 Table III.A.ii Estimated Mobile Wireless Coverage by Census Block Excluding Federal Land, Jan. 2014 Number of Providers with Coverage in a Block Number of Blocks POPs Contained in Those Blocks % of Total US POPs Square Miles Contained in Those Blocks % of Total US Square Miles Road Miles Contained in Those Blocks % of Total US Road Miles US Total 10,449,282 307,208,959 100.0 2,664,706 100.0 5,893,270 100.0 1 or more 10,335,706 306,912,383 99.9 2,260,521 84.8 5,764,976 97.8 2 or more 10,094,846 305,622,313 99.5 2,058,971 77.3 5,486,365 93.1 3 or more 9,194,561 298,173,820 97.1 1,561,830 58.6 4,612,503 78.3 4 or more 7,863,487 282,686,396 92.0 1,040,690 39.1 3,532,347 59.9 5 or more 4,106,624 147,056,170 47.9 456,311 17.1 1,712,159 29.1 Note: Based on January 2014 Mosaik Data and 2010 Census Data. The number of providers in a census block represent network coverage, which does not necessarily reflect the number of choices available to a particular individual or household, Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. Table III.A.iii Estimated Mobile Wireless Providers Offering Service by CMA, Excluding Territories, December 2011 Two Percent Market Share Threshold Five Percent Market Share Threshold Number of Providers Offering Service Anywhere in a CMA Number of CMAs Total CMAs (Percent) Number of CMAs Total CMAs (Percent) Total for U.S., excluding territories 716 100% 716 100% 1 provider 1 0.1% 2 0.3% 2 providers 51 7.1% 120 16.8% 3 providers 152 21.2% 213 29.7% 4 providers 210 29.3% 246 34.4% 5 or more providers 302 42.2% 135 18.9% Note: Market share analysis based on December 2011 NRUF data. The number of providers in a CMA which does not necessarily reflect the number of choices available to a particular individual or household. 119 Federal Communications Commission DA 14-1862 Table III.A.iv Estimated Mobile Wireless Broadband Coverage by Census Block Including Federal Land, Jan. 2014 Number of Providers with Coverage in a Block Number of Blocks POPs Contained in Those Blocks % of Total US POPs Square Miles Contained in Those Blocks % of Total US Square Miles Road Miles Contained in Those Blocks % of Total US Road Miles US Total 11,155,486 312,471,327 100.0 3,802,067 100.0 6,821,187 100.0 1 or more 10,791,991 311,491,813 99.7 2,669,327 70.2 6,322,249 92.7 2 or more 10,278,668 308,660,133 98.8 2,191,769 57.6 5,657,579 82.9 3 or more 8,502,584 291,761,257 93.4 1,284,356 33.8 4,013,320 58.8 4 or more 6,139,995 256,391,204 82.1 597,066 15.7 2,407,023 35.3 5 or more 1,008,800 36,863,284 11.8 94,732 2.5 393,493 5.8 Note: Based on January 2014 Mosaik Data and 2010 Census Data. The number of providers in a census block represent network coverage, which does not necessarily reflect the number of choices available to a particular individual or household, Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. Table III.A.v Estimated Mobile Wireless Broadband Coverage by Census Block Excluding Federal Land, Jan. 2014 Number of Providers with Coverage in a Block Number of Blocks POPs Contained in Those Blocks % of Total US POPs Square Miles Contained in Those Blocks % of Total US Square Miles Road Miles Contained in Those Blocks % of Total US Road Miles US Total 10,449,282 307,208,959 100.0 2,664,706 100.0 5,893,270 100.0 1 or more 10,230,158 306,455,948 99.8 2,173,496 81.6 5,648,813 95.9 2 or more 9,829,372 303,962,307 98.9 1,881,757 70.6 5,194,362 88.1 3 or more 8,244,196 288,200,564 93.8 1,172,461 44.0 3,817,550 64.8 4 or more 6,152,808 257,389,204 83.8 589,635 22.1 2,412,437 40.9 5 or more 2,571,728 114,527,905 37.3 169,321 6.4 870,215 14.8 Note: Based on January 2014 Mosaik Data and 2010 Census Data. The number of providers in a census block represent network coverage, which does not necessarily reflect the number of choices available to a particular individual or household, Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. 120 Federal Communications Commission DA 14-1862 Table III.A.vi Estimated Mobile Voice Coverage in Rural Areas by Census Block, Jan. 2014 Total Number of Providers with Coverage in a Block Number of Rural Census Blocks POPs Contained in Rural Census Blocks % of Total U.S. POPs Square Miles Contained in Those Blocks % of Total U.S. Square Miles Road Miles Contained in Those Blocks % of Total U.S. Road Miles Total for Rural U.S. 5,387,335 59,151,859 18.9 3,213,692 84.5 4,591,032 67.3 % of Total Rural U.S. POPs % of Total Rural U.S. Square Miles % of Total Rural U.S. Road Miles 1 or More 5,160,096 58,712,204 99.3 2,248,109 70.0 4,248,704 92.5 2 or More 4,779,873 57,001,226 96.4 1,825,852 56.8 3,726,980 81.2 3 or More 3,811,443 49,812,101 84.2 1,205,330 37.5 2,698,595 58.8 4 or More 2,555,860 37,625,516 63.6 650,851 20.3 1,612,597 35.1 5 or More 1,014,029 15,296,417 25.9 226,545 7.0 594,720 13.0 Note: Based on January 2014 Mosaik Data and 2010 Census Data. This table includes federal lands.The number of providers in a census block represent network coverage, which does not necessarily reflect the number of choices available to a particular individual or household, Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. Table III.A.vii Estimated Mobile Voice Coverage in Non-Rural Areas by Census Block, Jan. 2014 Total Number of Providers with Coverage in a Block Number of Non-Rural Census Blocks (Thousands) POPs Contained in Non-Rural Census Blocks (Thousands) % of Total U.S. POPs Square Miles Contained in Those Blocks (Thousands) % of Total U.S. Square Miles Road Miles Contained in Those Blocks (Thousands) % of Total U.S. Road Miles Total for Non-Rural U.S. 5,768,151 253,319,468 81.8 588,375 15.5 2,230,155 32.7 % of Total Non- Rural U.S. POPs % of Total Non- Rural U.S. Square Miles % of Total Non- Rural U.S. Road Miles 1 or More 5,750,963 253,260,183 100.0 546,881 92.9 2,213,491 99.3 2 or More 5,714,196 252,971,345 99.9 521,471 88.6 2,178,190 97.7 3 or More 5,593,620 251,393,015 99.2 466,532 79.3 2,079,909 93.3 4 or More 5,235,114 243,596,108 96.2 365,486 62.1 1,832,856 82.2 5 or More 1,364,861 52,683,798 20.8 103,069 17.5 498,572 22.4 Note: Based on January 2014 Mosaik Data and 2010 Census Data. This table includes federal lands.The number of providers 121 Federal Communications Commission DA 14-1862 in a census block represent network coverage, which does not necessarily reflect the number of choices available to a particular individual or household, Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. Table III.A.viii Estimated Mobile Broadband Coverage in Rural Areas by Census Block, Jan. 2014 Total Number of Providers with Coverage in a Block Number of Rural Census Blocks (Thousands) POPs Contained in Rural Census Blocks (Thousands) % of Total Rural U.S. POPs Square Miles Contained in Those Blocks (Thousands) % of Total Rural U.S. Square Miles Road Miles Contained in Those Blocks (Thousands) % of Total Rural U.S. Road Miles 1 or More 5,049,676 58,280,354 98.5 2,131,052 66.3 4,117,997 89.7 2 or More 4,581,100 55,844,770 94.4 1,681,950 52.3 3,497,094 76.2 3 or More 3,029,025 42,565,205 72.0 852,242 26.5 2,019,829 44.0 4 or More 1,394,545 23,427,622 39.6 303,882 9.5 813,212 17.7 5 or More 279,145 5,085,565 8.6 54,609 1.7 152,891 3.3 Note: Based on January 2014 Mosaik Data and 2010 Census Data. This table includes federal lands. The number of providers in a census block represent network coverage, which does not necessarily reflect the number of choices available to a particular individual or household, Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. Table III.A.ix Estimated Mobile Broadband Coverage in Non-Rural Areas by Census Block, Jan. 2014 Total Number of Providers with Coverage in a Block Number of Non-Rural Census Blocks (Thousands) POPs Contained in Non-Rural Census Blocks (Thousands) % of Total Non- Rural U.S. POPs Square Miles Contained in Those Blocks (Thousands) % of Total Non- Rural U.S. Square Miles Road Miles Contained in Those Blocks (Thousands) % of Total Non- Rural U.S. Road Miles 1 or More 5,742,315 253,211,459 100.0 538,275 91.5 2,204,252 98.8 2 or More 5,697,568 252,815,363 99.8 509,819 86.6 2,160,484 96.9 3 or More 5,473,559 249,196,052 98.4 432,114 73.4 1,993,491 89.4 4 or More 4,745,450 232,963,582 92.0 293,184 49.8 1,593,811 71.5 5 or More 729,655 31,777,719 12.5 40,123 6.8 240,602 10.8 Note: Based on January 2014 Mosaik Data and 2010 Census Data. This table includes federal lands. The number of providers in a census block represent network coverage, which does not necessarily reflect the number of choices available to a particular individual or household, Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. 122 Federal Communications Commission DA 14-1862 Chart III.A.i Table III.A.viii Estimated Mobile Wireless Coverage by Technology, Jan. 2014494 Technology POPs in Covered Blocks (Thousands) % of Total POPs Square Miles Contained in Those Blocks (Thousands) % of Total Square Miles Road Miles Contained in Those Blocks (Thousands) % of Total U.S. Road Miles CDMA 310,432 99.3 2,537 66.7 6,125 89.8 GSM/TDMA 310,454 99.4 2,520 66.3 6,091 89.3 iDEN 15,286 4.9 106 2.8 310 4.5 Total Digital 311,972 99.8 2,795 73.5 6,462 94.7 Note: Based on January 2014 Mosaik Data and 2010 Census Data. 494 Includes Federal lands. Commission estimates based on census block analysis of Mosaik CoverageRight coverage maps, October 2012. Population data are from the 2010 Census, and the square miles include the United States and Puerto Rico. 123 Federal Communications Commission DA 14-1862 Overall Connections and Customers Table III.B.i EA Penetration Rates 2013 Rank EA Market Name Penetration Rate 2012 2013 1 120 Grand Island, NE 101% 188% 2 57 Detroit-Ann Arbor-Flint, MI 128% 137% 3 55 Cleveland-Akron, OH-PA 121% 130% 4 122 Wichita, KS-OK 109% 127% 5 51 Columbus, OH 111% 120% 6 83 New Orleans, LA-MS 118% 116% 7 20 Norfolk-Virginia Beach-Newport News, VA-NC 111% 115% 8 89 Monroe, LA 123% 115% 9 10 New York-North New Jersey-Long Island, NY- NJ-CT-PA 111% 115% 10 13 Washington-Baltimore, DC-MD-VA-WV-PA 117% 114% 11 85 Lafayette, LA 113% 114% 12 49 Cincinnati-Hamilton, OH-KY-IN 112% 114% 13 64 Chicago-Gary-Kenosha, IL-IN-WI 108% 114% 14 135 Odessa-Midland, TX 110% 114% 15 111 Minot, ND 113% 113% 16 90 Little Rock-North Little Rock, AR 117% 113% 17 31 Miami-Fort Lauderdale, FL 108% 112% 18 50 Dayton-Springfield, OH 106% 112% 19 84 Baton Rouge, LA-MS 111% 111% 20 155 Farmington, NM-CO 110% 111% 21 87 Beaumont-Port Arthur, TX 108% 111% 22 40 Atlanta, GA-AL-NC 109% 111% 23 17 Roanoke, VA-NC-WV 110% 110% 24 3 Boston-Worcester-Lawrence-Lowewell- Brockton, MA-NH 107% 110% 25 97 Springfield, IL-MO 108% 109% 26 171 Anchorage, AK 98% 109% 27 73 Memphis, TN-AR-MS-KY 107% 109% 28 12 Philadelphia-Wilmington-Atlantic City, PA-NJ- DE-MD 106% 109% 29 99 Kansas City, MO-KS 104% 109% 30 44 Knoxville, TN 108% 109% 31 71 Nashville, TN-KY 111% 108% 32 88 Shreveport-Bossier City, LA-AR 112% 108% 33 124 Tulsa, OK-KS 118% 108% 124 Federal Communications Commission DA 14-1862 34 79 Montgomery, AL 106% 108% 35 22 Fayetteville, NC 110% 107% 36 15 Richmond-Petersburg, VA 106% 107% 37 34 Tampa-St. Petersburg-Clearwater, FL 104% 107% 38 78 Birmingham, AL 107% 107% 39 37 Albany, GA 104% 106% 40 131 Houston-Galveston-Brazoria, TX 104% 106% 41 80 Mobile, AL 105% 106% 42 132 Corpus Christi, TX 103% 106% 43 172 Honolulu, HI 102% 106% 44 86 Lake Charles, LA 109% 106% 45 93 Joplin, MO-KS-OK 104% 106% 46 125 Oklahoma City, OK 115% 106% 47 127 Dallas-Fort Worth, TX-AR-OK 103% 106% 48 81 Pensacola, FL 104% 106% 49 77 Jackson, MS-AL-LA 103% 106% 50 29 Jacksonville, FL-GA 102% 105% 51 56 Toledo, OH 102% 105% 52 96 St. Louis, MO-IL 105% 105% 53 141 Denver-Boulder-Greeley, CO-KS-NE 103% 105% 54 69 Evansville-Henderson, IN-KY-IL 99% 105% 55 107 Minneapolis-St. Paul, MN-WI-IA 102% 105% 56 152 Salt Lake City-Ogden, UT-ID 101% 105% 57 82 Biloxi-Gulfport-Pascagoula, MS 105% 105% 58 161 San Diego, CA 103% 105% 59 53 Pittsburgh, PA-WV 103% 105% 60 38 Macon, GA 105% 105% 61 170 Seattle-Tacoma-Bremerton, WA 103% 104% 62 163 San Francisco-Oakland-San Jose, CA 101% 104% 63 24 Columbia, SC 100% 104% 64 74 Huntsville, AL-TN 106% 104% 65 63 Milwaukee-Racine, WI 99% 104% 66 70 Louisville, KY-IN 101% 104% 67 27 Augusta-Aiken, GA-SC 104% 104% 68 8 Buffalo-Niagara Falls, NY-PA 101% 104% 69 153 Las Vegas, NV-AZ-UT 101% 104% 70 134 San Antonio, TX 99% 103% 71 41 Greenville-Spartanburg-Anderson, SC-NC 100% 103% 72 128 Abilene, TX 98% 102% 73 160 Los Angeles-Riverside-Orange County, CA-AZ 100% 102% 74 45 Johnson City-Kingsport-Bristol, TN-VA 103% 102% 75 23 Charlotte-Gastonia-Rock Hill, NC-SC 101% 102% 125 Federal Communications Commission DA 14-1862 76 159 Tucson, AZ 97% 102% 77 35 Tallahassee, FL-GA 100% 102% 78 18 Greensboro-Winston-Salem-High Point, NC-VA 101% 101% 79 5 Albany-Schenectady-Troy, NY 100% 101% 80 136 Hobbs, NM-TX 85% 101% 81 101 Peoria-Pekin, IL 98% 100% 82 30 Orlando, FL 97% 100% 83 67 Indianapolis, IN-IL 98% 100% 84 95 Jonesboro, AR-MO 105% 100% 85 143 Casper, WY-ID-UT 98% 100% 86 16 Staunton, VA-WV 99% 100% 87 2 Portland, ME 98% 100% 88 43 Chattanooga, TN-GA 99% 99% 89 137 Lubbock, TX 96% 99% 90 130 Austin-San Marcos, TX 96% 99% 91 133 McAllen-Edinburg-Mission, TX 97% 99% 92 7 Rochester, NY-PA 97% 99% 93 42 Asheville, NC 98% 99% 94 75 Tupelo, MS-AL-TN 97% 99% 95 103 Cedar Rapids, IA 100% 99% 96 102 Davenport-Moline-Rock Island, IA-IL 98% 99% 97 154 Flagstaff, AZ-UT 97% 99% 98 25 Wilmington, NC-SC 96% 98% 99 142 Scottsbluff, NE-WY 98% 98% 100 158 Phoenix-Mesa, AZ-NM 95% 98% 101 129 San Angelo, TX 94% 98% 102 26 Charleston-North Charleston, SC 98% 98% 103 9 State College, PA 96% 98% 104 138 Amarillo, TX-NM 93% 98% 105 28 Savannah, GA-SC 97% 98% 106 6 Syracuse, NY-PA 96% 98% 107 72 Paducah, KY-IL 93% 98% 108 167 Portland-Salem, OR-WA 95% 98% 109 66 Fort Wayne, IN 95% 97% 110 62 Grand Rapids-Muskegon-Holland, MI 93% 97% 111 76 Greenville, MS 97% 97% 112 59 Green Bay, WI-MI 94% 97% 113 106 Rochester, MN-IA-WI 95% 97% 114 36 Dothan, AL-FL-GA 93% 97% 115 39 Columbus, GA-AL 100% 97% 116 48 Charleston, WV-KY-OH 99% 97% 117 109 Duluth-Superior, MN-WI 93% 97% 126 Federal Communications Commission DA 14-1862 118 19 Raleigh-Durham-Chapel Hill, NC 96% 96% 119 100 Des Moines, IA-IL-MO 96% 96% 120 119 Lincoln, NE 93% 96% 121 91 Fort Smith, AR-OK 99% 96% 122 94 Springfield, MO 90% 96% 123 151 Reno, NV-CA 92% 96% 124 144 Billings, MT-WY 93% 96% 125 118 Omaha, NE-IA-MO 95% 96% 126 11 Harrisburg-Lebanon-Carlisle, PA 93% 96% 127 164 Sacramento-Yolo, CA 92% 96% 128 148 Idaho Falls, ID-WY 95% 95% 129 166 Eugene-Springfield, OR-CA 92% 95% 130 157 El Paso, TX-NM 92% 95% 131 139 Santa Fe, NM 94% 95% 132 4 Burlington, VT-NY 92% 95% 133 156 Albuquerque, NM-AZ 93% 95% 134 98 Columbia, MO 95% 95% 135 52 Wheeling, WV-OH 95% 95% 136 110 Grand Forks, ND-MN 93% 94% 137 116 Sioux Falls, SD-IA-MN-NE 93% 94% 138 169 Richland-Kennewick-Pasco, WA 92% 94% 139 147 Spokane, WA-ID 92% 94% 140 126 Western Oklahoma, OK 102% 94% 141 68 Champaign-Urbana, IL 92% 93% 142 32 Fort Myers-Cape Coral, FL 91% 93% 143 165 Redding, CA-OR 90% 93% 144 65 Elkhart-Goshen, IN-MI 90% 93% 145 123 Topeka, KS 90% 93% 146 140 Pueblo, CO-NM 89% 93% 147 149 Twin Falls, ID 92% 92% 148 1 Bangor, ME 93% 92% 149 113 Fargo-Moorhead, ND-MN 90% 92% 150 54 Erie, PA 90% 92% 151 33 Sarasota-Bradenton, FL 90% 92% 152 60 Appleton-Oshkosh-Neenah, WI 90% 92% 153 150 Boise City, ID-OR 89% 92% 154 108 Wausau, WI 92% 92% 155 117 Sioux City, IA-NE-SD 91% 92% 156 104 Madison, WI-IA-IL 90% 91% 157 47 Lexington, KY-TN-VA-WV 88% 90% 158 21 Greenville, NC 90% 90% 159 145 Great Falls, MT 88% 90% 127 Federal Communications Commission DA 14-1862 160 46 Hickory-Morganton, NC-TN 90% 89% 161 14 Salisbury, MD-DE-VA 90% 89% 162 162 Fresno, CA 84% 87% 163 115 Rapid City, SD-MT-ND-NE 90% 87% 164 92 Fayetteville-Springdale-Rogers, AR-MO-OK 88% 86% 165 168 Pendleton, OR-WA 83% 86% 166 105 La Crosse, WI-MN 84% 85% 112 Bismarck, ND-MT-SD 94% * 58 Northern Michigan, MI * * 61 Traverse City, MI * * 114 Aberdeen, SD * * 121 North Platte, NE-CO * * 146 Missoula, MT * * Consumers and Mobile Wireless Table III.C.i Market Share by Smartphone Model, 2009 – 2013 Operating System Developer Share of Smartphones in Use December 2009 August 2010 September 2011 September 2012 September 2013 Google/Android 5.2% 19.6% 44.8% 52.2% 51.80% Apple 25.3% 24.2% 27.4% 34.3% 40.60% RIM/BlackBerry 41.6% 37.6% 18.9% 8.4% 3.8% Microsoft 18.0% 10.8% 5.6% 3.6% 3.3% Palm 6.1% 4.6% 0.0% 0.0% 0.0% Symbian 0.0% 0.0% 1.8% 0.6% 0.3% All Others 3.8% 3.2% 1.5% 0.6% 0.20% Note: based on ComScore MobiLens 3-month survey data averages 128 Federal Communications Commission DA 14-1862 Table III.C.ii Percentage of U.S. Adults Living in Households with/without Wireless and Landlines (2008 - 2013) Percent of Adults in Households with: Date of interview Landline with Wireless Landline without Wireless Wireless- only Phoneless Jul–Dec 2008 63.7% 15.1% 18.4% 1.7% Jan–Jun 2009 63.5% 13.4% 21.1% 1.5% Jul–Dec 2009 62.5% 12.6% 22.9% 1.7% Jan–Jun 2010 62.2% 10.9% 24.9% 1.7% Jul–Dec 2010 59.4% 10.7% 27.8% 1.8% Jan–Jun 2011 58.8% 9.0% 30.2% 1.8% Jul–Dec 2011 57.3% 8.3% 32.3% 1.9% Jan–Jun 2012 56.1% 7.8% 34.0% 1.9% Jul–Dec 2012 54.4% 7.0% 36.5% 1.9% Jan–Jun 2013 52.8% 6.9% 38.0% 2.2% Jun-Dec 2013 51.5% 7.0% 39.1% 2.2% Note: Adults are aged 18 and over, children are under age 18, Source: CDC/NCHS National Health Interview Survey Early Release Program, “Wireless Substitution: Early Release of Estimates From the National Health Interview Survey, July–December 2013”, Table 1, Released July 2014 Table II.C.iii Percentage of U.S. Children Living in Households with/without Wireless and Landlines (2008 - 2013) Percent of Children in Households with: Date of interview Landline with Wireless Landline without Wireless Wireless- only Phoneless Jul–Dec 2008 67.1% 11.1% 18.7% 2.4% Jan–Jun 2009 67.6% 9.1% 21.3% 1.7% Jul–Dec 2009 63.4% 8.5% 25.9% 1.9% Jan–Jun 2010 62.8% 6.4% 29.0% 1.7% Jul–Dec 2010 59.8% 6.2% 31.8% 2.0% Jan–Jun 2011 56.7% 5.1% 36.4% 1.7% Jul–Dec 2011 54.7% 4.8% 38.1% 2.2% Jan–Jun 2012 52.7% 4.5% 40.6% 2.2% Jul–Dec 2012 49.5% 3.4% 45.0% 1.9% Jan–Jun 2013 48.3% 3.6% 45.4% 2.6% Jul–Dec 2014 46.4% 3.8% 47.1% 2.5% Note: Adults are aged 18 and over, children are under age 18, Source: CDC/NCHS National Health Interview Survey Early Release Program, “Wireless Substitution: Early Release of Estimates From the National Health Interview Survey, July–December 2013”, Table 1, Released July 2014. 129 Federal Communications Commission DA 14-1862 APPENDIX IV INPUT MARKET Non-Spectrum Input Segments Table IV.B.i Year End Cell Site Counts by Provider, 2011 – 2014495 Note: Cell site counts for individual service provider are from UBS Wireless 411: Version 54, Figure 48. The total industry-wide cell count is from CTIA2013, at 114, June 2014. 495 See CTIA, 2013 Annual Wireless Industry Survey Results (“CTIA2013”), at 114, June 2014. Because multiple cell sites can be co-located in the same “tower” site, the reported cell sites should not be equated with “towers.” See also CTIA2013 at 105. The reported cell sites include repeaters and other cell-extending devices (e.g., femtocells, or distributed antenna systems). See CTIA2013 at 105 and 106. Cell Sites 2011 2012 2013 1st Half 2014 Verizon Wireless 43,390 44,590 46,655 47,855 AT&T 56,200 56,900 61,800 63,700 Sprint 67,500 57,900 55,000 55,000 T-Mobile 50,545 51,104 63,879 63,945 Leap 9,000 9,000 9,000 NTELOS 1,353 1,429 1,444 1,445 US Cellular 7,882 8,028 6,975 6,183 Total by Top Seven Reported Service Providers 235,870 228,951 244,753 238,128 CTIA Reported Total Industry- wide Cell Sites 283,385 301,779 304,360 130 Federal Communications Commission DA 14-1862 APPENDIX V PRICING LEVELS AND TRENDS Postpaid Service Table V.A.i Equipment Installment Payment (EIP) Plans with Early Upgrade Option Date Provider Plan Name Description 7/13 T-Mobile Jump “Un-carrier 2.0” Customers can upgrade their device every six months for an additional $10 per month on top of an existing plan with trade-in of existing device.i 7/13 AT&T Next Customers can buy a device and upgrade to a new device every 12 months with no down payment and trade-in of existing device.ii 7/13 Verizon Edge When Edge was first introduced, customers were allowed to upgrade every 6 months if 50 percent of the retail cost of the handset had been paid off,iii Share Everything data plan required. 9/13 to 1/14 Sprint One Up Customers can purchase a device with no down payment and upgrade after 12 months with trade-in of existing device.iv 1/14 Sprint Framily Plans/ Easy Pay Customers can get a smartphone or feature phone for a down payment and 24 monthly installment payments and upgrade after 12 months on certain qualifying plans with trade-in of existing or equivalent model.v Early upgrade option (annual upgrades) available only with purchase of unlimited data plan.vi 1/14 Verizon Edge Customers who enroll can select a device and then upgrade after 30 days with trade-in of existing device if at least 50% of the retail price of the device is paid off.vii Share Everything data plan required. Promotion allowing Edge customers to upgrade after 30 days.viii 2/14 T-Mobile Jump Customers can upgrade any time they want and as often as they want if they have paid off at least 50% of the cost of the device (T-Mobile will cover remaining payments up to half of device cost). Trade-in requirement and $10 monthly fee still apply. Tablets added to Jump plan.ix 6/14 Verizon Edge Changes to conditions of upgrade program: x (1) Customers are required to pay off 60% of the retail price of the device before upgrading, up from 50% previously. (2) Device financing costs are spread over 20 months instead of the previous 24 months, thereby raising customers’ monthly equipment installment payments but allowing them to pay off devices more rapidly. 131 Federal Communications Commission DA 14-1862 Table V.A.ii Discounts on Monthly Service Fees for EIP Customers496 Date Provider Plan Name Description 9/13 - 1/14 Sprint One Up $15/month discount per line off unlimited contract plans (Unlimited, My Way and My All-in).xi 12/13 AT&T Mobile Share Value/ Next $15/month discount per line – from $40/month per device to $25/month per device.xii 1/14 Sprint Framily Plans/ Easy Pay Gives customers progressively larger discounts as more people join the Framily up to a maximum discount of $30/month per line and a limit of 10 phone numbers per group.xiii $5/month discount per line off unlimited data contract plan; $15/month discount per line off 1GB data contract plan. All Framily plans include unlimited voice and text messaging, which starts at $55/month for one line and progressively decreases in price per line as more lines are added. Options for data add-ons - unlimited data for $20/month, 3GB for $10/month and 1GB for no additional charge. The 3GB and 1GB data add-ons have an overage charge of 1.5 cents/MB.xiv 2/14 AT&T Mobile Share Value/ Next Reduces price by $40 on a plan with four smartphones with unlimited voice and text and 10 GB of data which now costs $160. Additional $10/month discount per line for Next customers with data buckets of 10GB or more – from already discounted $25/month per device to $15/month per device.xv 2/14 Verizon More Everything/ Edge $10/month discount per line off smartphone access for data buckets of 8GB or less, and $20/month discount per line off smartphone access for data buckets of 10GB or more – from $40/per month per line to $30/month per line for 8GB bucket or less, and $20/month per line for 10GB bucket or more. Discounts limited to Edge customers who pay for devices with EIP.xvi 4/14 Verizon More Everything $25/month discount per line off smartphone access for data buckets of 10GB or more – from already discounted $20/month per line to $15/month per line.xvii Matches AT&T in offering 10GB for 4 lines for $160/month. 4/14 C Spire Unlimited Everything $35/month discount off existing unlimited voice and data plan – from $100/month to $65/month. Plan includes unlimited voice, texting, picture messaging, web, music and video. xviii 4/14 Verizon More Everything Eligibility for EIP discounts on monthly smartphone access fees extended to existing customers on month-to-month contracts and new customers who bring their own device -- $30/month per line 496 T-Mobile, AT&T and Sprint have consistently offered the same EIP discounts to customers who pay the full retail price of the device upfront, bring their own device, and those who have already paid for their device under a traditional two-year service contract that has expired. In contrast, when Verizon first started offering EIP discounts, the company initially limited eligibility to Edge customers. Subsequently, Verizon extended the same discounts to existing customers on month-to-month contracts and new customers who bring their own device, but only for a limited time. Phil Goldstein, Analysts: Verizon’s ‘More Everything’ Plans Show No. 1 Carrier Isn’t Immune to Price War, FIERCEWIRELESS, Feb. 13, 2014; Debi Lewis, More Savings Coming for Verizon Wireless Customers Beginning April 17, Press Release, Verizon Wireless, Apr. 14, 2014. 132 Federal Communications Commission DA 14-1862 to add a smartphone for data buckets of 8GB or less, and $15/month per line for data buckets of 10 GB or more. The new offer is available only for a limited time.xix 5/14 C Spire Shared Data $15/month discount per line off smartphone access for data buckets of 8 GB or less – from $40/month per line to $25/month per line.xx $25/month discount per line off smartphone access for data buckets of 10GB or more – from $40/month per line to $15/month per line. Matches AT&T and Verizon in offering 10GB for 4 lines for $160/month.xxi Table V.A.iii Major Changes to Pricing of Postpaid Service Plans Date Provider Plan Name Description 3/13 T-Mobile Simple Choice “Un-carrier 1.0” Discontinued offering of traditional two-year contracts with monthly fees that include the cost of device subsidies discontinued. All new retail customers must sign up for no-contract plans that separate service and equipment fees. EIP option available to customers with good credit. xxiiixxii 7/13 Sprint Unlimited, My Way; My All-in Launched new unlimited rate plans – All plans include unlimited voice and text, which starts at $50/month for one line. As more devices are added to account, customers get progressively larger discounts, up to a maximum discount of $30/month per line and a limit of 10 lines per account. Customers can also purchase 1GB per month of mobile hotspot usage for $10/month. “My All-in” also includes 5GB of mobile hotspot usage for a total of $110/month. xxiv Unlimited data for smartphones is offered as a $30/month add-on. Therefore, unlimited voice, text and data for one device costs $80/month, down $30/month from previous Simply Everything unlimited rate plan. Launched 1GB data tier - option is offered as a $20/month add-on. - $70/month including unlimited voice and text.xxv Sprint becomes last nationwide operator to shift from exclusively unlimited data pricing to tiered, usage-based data pricing for smartphones. 10/13 AT&T Mobile Share Discontinued offering of traditional voice and data plans based on usage of individual devices. All new retail customers must sign up for shared data plans.xxvi 12/13 AT&T Mobile Share Restructured rate plan -- standard $40/month fee to add a smartphone line replaces previous sliding scale of $30-50/month depending on data bucket size. Price of 1-2GB buckets raised $5/month, and price of 6-20GB buckets reduced by $10-50/month. Before the change, the cost to add a smartphone to a shared data plan was $50/month for the 300MB tier, $45/month for 1-2GB, $40/month for 4GB, $35/month for 6GB and $30/month for larger data tiers.xxvii xxviii Net effect on monthly service fees depends on usage and number of smartphone lines. 133 Federal Communications Commission DA 14-1862 1/14 Verizon Share Everything Introduced less expensive entry-level plan with smaller data bucket: $60/month for 250MB of smartphone data.xxix 2/14 AT&T Mobile Share Restructured rate plan and offered to existing AT&T customers -- monthly line access fees reduced for existing subsidized contract customers until next upgrade -- from $40/month per device to $25/month for data buckets of 2GB to 6GB, and to $15/month for data buckets of 10GB or more. The change effectively allowed current traditional contract customers with subsidized devices to switch to the same discounted pricing structure as AT&T’s Mobile Share Value plans for Next/ EIP customersxxx Current customers can keep the discounted pricing indefinitely with their existing devices. However, once customers decide to upgrade to a new device, they will have to sign up for the Next EIP plan or purchase the new device outright to keep the discounted pricing. Prices will go back up if they choose to get another subsidized device through a traditional contract plan. 2/14 Verizon Loyalty Plans Loyalty plans offered to retain customers: $60/per month for unlimited voice, texting and 2GB of smartphone data.xxxi 2/14 Verizon More Everything Increased data allowances on selected tiers without increasing monthly prices – from 500MB to 1GB on $40/month plan, from 1GB to 2GB on $50/month plan and from 2GB to 3GB on $60/month plan. Unlimited international messaging and 25GB of cloud storage per line added to all pricing tiers.xxxii The monthly fee on basic 700 MOU voice-only calling plan was also reduced by $5/month -- from $40/ month to $35/month. Monthly data fees reduced by $10/month on 500MB data bucket (), and by $5/month on 250MB data bucket (). 3/14 T- Mobile Simple Choice Increased high-speed data allowances on tiered data plans without increasing monthly prices – from 500MB to 1GB on $50/month entry-level plan, and from 2.5GB to 3GB on $60/month plan. Unlimited international texting added to all tiers.xxxiii Monthly price of unlimited data plan increased $10/month, from $70/month to $80/month. Tethered data allowance on unlimited plan doubled to 5GB. New 5GB high-speed data tier introduced at $70/month price point of original unlimited data plan.xxxiv 3/14 AT&T Mobile Share/ Mobile Share Value Monthly price reduction of entry-level 2GB data bucket for single- line customers and for two lines sharing the 2GB data allowance -- from $55/month to $40/month (excluding monthly device access fees per line).xxxv Price cut applies equally to EIP and subsidized subscribers. Unlimited international messaging added to all tiers at no additional cost. 3/14 U.S. Cellular Data allowances increased on tiered plans without increasing monthly prices – from 300MB to 1GB on $40/month plan, from 1GB to 2GB on $50/month plan, and from 2GB to 3GB on $60/month plan.xxxvi 4/14 T-Mobile Simple Starter497 Introduction of new entry-level 500MB LTE data plan for $40/month, with service suspension for the month after LTE data 497 In place of overage charges or speed reductions, the service is suspended entirely for the month once the user reaches the 500MB LTE data allowance. Customers who wish to exceed the 500MB LTE data allowance need to purchase either a 134 Federal Communications Commission DA 14-1862 allowance reached. In place of overage charges or speed reductions, the service is suspended entirely for the month once the user reaches the 500MB LTE data allowance. Customers who wish to exceed the 500MB LTE data allowance need to purchase either a 500MB one-day on-network data pass for $5 or a 1GB 7-day on- network data pass for $10. The plan does not include international service packages or roaming. xxxvii 4/14 T-Mobile All plans Overage charges abolished on all plans, including legacy plans, for domestic calls, text messages and data usage.xxxviii 8/14 Sprint Family Share Pack New shared-data plan that offers double the high-speed data at a lower price than AT&T and Verizon Wireless. $100 for 4 lines with 28 GB of shared data and $100 for 10 lines and 40 GB of shared data. % GB of data per line plus 2 GB of extra data per line through 2015.xxxix 12/14 T-Mobile Simple Choice Introduces unlimited 4G LTE Family Plan for $100 for 2 lines and goes all the way up to 10 people for $40 more per line. xl 500MB one-day on-network data pass for $5 or a 1GB 7-day on-network data pass for $10. The plan does not include international service packages or roaming. 135 Federal Communications Commission DA 14-1862 Table V.A.iv Selected Promotions and Incentives to Attract and Retain Customers Date Provider Plan Name Description 7/13 Sprint Unlimited, My Way; My All-in Guarantees unlimited data for the life of the line to new customers who sign up for, and existing customers who switch to, Unlimited, My Way or My All-in plans.xli 10/13 AT&T New tablet data plans: $5/day plan with 250MB per day, and $25 plan with 1GB that can be used any time during 3- month period.xlii 10/13 T-Mobile “Un-carrier 4.0” Tablets Un-leashed All customers who use tablets on the network have lifetime access to 200MB of free LTE data per month. Customers need to purchase a $10 SIM card from T-Mobile to connect a tablet. Existing voice customers with a phone plan and new customers who pay full price for tablets can access the 200MB of free data without signing up for additional plans, but customers who wish to pay for a tablet using the company’s no-money-down EIP are required to purchase a monthly data plan before gaining access to the 200MB of free data.xliii 1/14 Sprint Framily Plan Customers who join a Framily plan can get a free Galaxy Tab 3 tablet if they sign up for a qualifying data plan. Sprint’s free tablet offer began as an initially short-term promotion in January 2014, but was extended multiple times during the first quarter of 2014. xliv 1/14 to 2/14 AT&T Customers who switch to AT&T from T-Mobile and trade-in their eligible smartphone can get up to $450. When they pay for their device (Next EIP, full retail price or BYOD) get $200 credit per line and ability to trade in current device for promotion card worth up to $250, which could be used toward purchase of AT&T products or services.xlv 1/14 T-Mobile Offer to reimburse up to $650 in ETFs for customers switching to T-Mobile from the other 3 nationwide providers (AT&T, Verizon or Sprint). Offer includes up to $300 device credit for trading in handset and up to $350 payment per line depending on proof of ETF paid to previous provider. In order to qualify, a customer must trade in their old handset, purchase a new T-Mobile handset, and port their phone number to T-Mobile. xlvi 2/14 to 3/14 AT&T Mobile Share New and existing customers who open new line of service will receive a $100 credit for each new smartphone, feature phone, tablet, mobile Hot Spot or Wireless Home Phone they add.xlvii 2/10/14 to 2/17/14 Verizon No-activation fee promotion: $35 activation fee waived for new customers who signed a two-year service contract during the promotion period.xlviii 4/14 C Spire Customers who switch to C Spire get up to $200 service credit as reimbursement for ETFs, plus $50 bill credit and an extra $100 for every third customer referral.xlix 136 Federal Communications Commission DA 14-1862 4/14 Verizon More Everything Customers who activate a tablet on shared data plans at or above the 1GB data tier get 1GB of extra data per month free of additional charge. There is a $10/month charge for adding a tablet line to shared data plans.l 4/14 to 5/14 Sprint Framily Offer to reimburse up to $650 in switching costs for customers who switch number to Sprint from another postpaid provider and sign up for Sprint Framily Plan. Offer includes up to $300 credit for phone trade-in and a Visa® Prepaid Card worth up to $350 to cover ETFs or, for customers not on contract, their remaining equipment installment balance.li 4/14 to ? U.S. Cellular Shared Connect Offer to pay off customers’ old contracts, up to $350 per line, when they switch to Shared Connect plan and choose Retail Installment Contract. Customers who port their number from another provider and send in their final bill with the ETF on it will receive a prepaid debit card for that amount.lii 4/14 to 5/14 T-Mobile LTE-enabled tablets offered for same price as Wi-Fi-only models for any postpaid activation on a 1GB or more mobile internet data plan.liii 4/14 T-Mobile $10/month discount off most popular internet data plans for new and existing customers through end of 2014.liv 6/14 T-Mobile T-Mobile Test Drive Un-carrier 5.0 From June 23, customers will be able to test iPhone 5s with unlimited service for seven days at no cost.lv 12/5 Sprint Cut Your Bill in Half Event Effective 12/05/2014, available for limited time only, Sprint is introducing the Cut Your Bill in Half Event for Verizon and AT&T customers who are interested in switching to Sprint to cut their rate plan in half. 137 Federal Communications Commission DA 14-1862 Table V.A.v Selected Postpaid Plans for Basic and Smartphones as of July 2014 Provider Phone type Plan Name Talk/Text Sample Data Tier Details Voice and Text Data Voice + Data + Text AT&T Basic/ Feature Phone AT&T Mobile Share® Value Plans with Unlimited Talk & Text Unlimited 300MB Base fee for basic phone=$20, until the data reaches 10 GB and it falls to $15. Additional basic or smartphones may be added. Data tiers are purchased for the following prices: 300MB=$20; 1GB=$25; 2GB=$40; 4GB=$70; 6GB=$80; 10GB=$100; 15GB=$130; 20GB=$150; 30GB= $225; 40GB=$300; 50GB=$375 $ 20.00 $ 20.00 $ 40.00 AT&T Smartphone AT&T Mobile Share® Value Plans with Unlimited Talk & Text Unlimited 1 GB Base fee for smartphone = $40. Additional basic or smartphones may be added. Data tiers are purchased for the following prices: 300MB=$20; 1GB=$25; 2GB=$40; 4GB=$70; 6GB=$80; 10GB=$100; 15GB=$130; 20GB=$150; 30GB=$225; 40GB=$300; 50GB=$375 $ 40.00 $ 25.00 $ 65.00 Verizon Wireless Basic/ Feature Phone More Everything Plan Unlimited 700 min Base fee=$30. Data added [how much data?] for $5 $ 30.00 $ 5.00 $ 35.00 Verizon Wireless Smartphone More Everything Plan Unlimited 1GB Base fee = $40. Data tiers are purchased for the following prices: 250MB=$15; 500MB=$30; 1GB=$40; 2GB=$50; 3GB=$60; 4GB=$70; 6GB=$80; 8GB=$90; 10GB=$100; 12GB=$110; 14GB=$120; 16GB=$130; 18GB=$140; 20GB=$150; 30GB=$225; 40GB=$300; 50GB=$375 $ 40.00 $ 40.00 $ 80.00 138 Federal Communications Commission DA 14-1862 Provider Phone type Plan Name Talk/Text Sample Data Tier Details Voice and Text Data Voice + Data + Text Sprint Basic/ Feature Phone Unlimited My Way Unlimited Unlimited Base fee = $50. Unlimited data for basic phones can be added for $10. $ 50.00 $ 10.00 $ 60.00 Sprint Smartphone Unlimited My Way Unlimited 1 GB Base fee = $50. Data tiers are purchased for the following prices: 1GB=$20; unlimited data=$30 $ 50.00 $ 20.00 $ 70.00 Sprint Basic/ Feature Phone Framily Plan Unlimited 1 GB Base fee for initial phone = $55. Per- phone fee of $55 falls by $5 for every member of plan as new members are added, until it reaches $25. The maximum size of a framily is 10 people. Plan comes with 1GB data. Higher data tiers may also be purchased. $ 55.00 $ - $ 55.00 Sprint Smartphone Framily Plan Unlimited 3 GB Base fee for initial phone = $55. Per- phone fee of $55 falls by $5 for every member of plan as new members are added, until it reaches $25. The maximum size of a framily is 10 people. Plan comes with 1GB data. Higher data tiers may also be purchased: 3GB=$10, unlimited data=$20 $ 55.00 $ 10.00 $ 65.00 T- Mobile Basic/ Feature Phone Simple Starter Plan Unlimited 500 MB Plan may be used with smartphones or basic phones $ 40.00 $ - $ 40.00 T- Mobile Smartphone Simple Choice Plan Unlimited 1 GB Base fee = $50. Data tiers are purchased for the following prices: 1GB=included with plan; 3GB=$10; 5GB=$20; unlimited data=$30 $ 50.00 $ - $ 50.00 139 Federal Communications Commission DA 14-1862 Provider Phone type Plan Name Talk/Text Sample Data Tier Details Voice and Text Data Voice + Data + Text US Cellular Basic/ Feature Phone Shared Connect Plans Unlimited 300 MB Base fee=$30. Data tiers are purchased for the following prices: 300 MB=$15, 1GB=$40; 2 GB=$50; 3GB=$60 $30 $40 $70 US Cellular Smartphone Shared Connect Plans Unlimited 1 GB Base fee=$40. Data tiers are purchased for the following prices: 300 MB=$15, 1GB=$40; 2 GB=$50; 3GB=$60 $40 $40 $80 C Spire Basic/ Feature Phone Unlimited talk & text Unlimited (including pictures) Unlimited (assume 30 days) $1 for 24 hours of data access $50 $30 $80 C Spire Smartphone Unlimited Everything Plans Unlimited (including pictures) Unlimited Unlimited $80 C Spire Smartphone Unlimited talk & text Unlimited (including pictures) 1 month of streaming Base fee=$50. Data passes available ($5 for 2 hrs., $10 for 5 hrs., $30 for 1 month of streaming) $50 $30 $80 C Spire Smartphone Unlimited talk & text Unlimited (including pictures) 1 GB Base fee=$50. No data included. Data tiers are purchased for the following prices (valid 30 days): 500 MB =$10, 1GB=$15; 3 GB=$45; 5 GB= $75 $50 $15 $65 C Spire Smartphone Plans with second Line Discount: Unlimited Unlimited (including pictures) Unlimited Can add a second line on the plan for $50 $100 $- $100 C Spire Smartphone Plans with second Line Discount: Unlimited Lite Unlimited (including pictures) Unlimited web and online music, 30 min. of online video Can add a second line on the plan for $50. Data passes available ($5 for 2 hrs., $10 for 5 hrs., $30 for 1 month of streaming) $80 $- $80 140 Prepaid Service Table V.B.i Selected Major Developments in Prepaid Pricing Plans Date Provider Brand/MVNO Description 5/13 AT&T Aio New brand launched on HSPA+ network in selected regional markets. Plans for feature phones and smartphones include unlimited voice and text and a choice of data tiers, with speed reductions after data allowances reached.lvi Tablet plans also available. 5/13 T-Mobile MetroPCS After T-Mobile-MetroPCS transaction closes, combined company begins process of migrating customers off MetroPCS legacy CDMA network onto LTE network and taking MetroPCS brand nationwide.lvii 5/13 Verizon Data allowances increased on prepaid plans: unlimited voice, text and 2GB of data for $60/month, up from 500MB of data; unlimited voice, text and 4GB of data for $70/month, up from 2GB of data.lviii 6/13 AT&T Aio LTE service added to service originally launched with support for only HSPA+ network. Service footprint also expanded to additional markets.lix 7/13 T-Mobile MetroPCS New entry-level rate plan added: $40/month for unlimited voice, text and 500MB of LTE data, with data speeds slowed after data allowance is reached.lx 8/13 AT&T Aio Service launched on nationwide basis.lxi Positioned as competitor to prepaid brands including Sprint’s Boost and Virgin Mobile brands and T- Mobile USA’s MetroPCS brand.lxii 10/13 T-Mobile Brightspot New prepaid offering launched with Target as retail partner.lxiii 2/14 to 3/14 Sprint Boost LTE promotion cuts price of unlimited voice, text and 2.5GB of data before speed reductions from $55/month to $35/month for first 6 months, after which price will increase to $50/month. Once introductory period ends, customers will qualify for Boost’s Shrinking Payments plan, under which Sprint will take $5 off the customer’s monthly bill for every six on-time payments, down to a floor of $40/month.lxiv 3/14 Verizon Allset New plans for 3G CDMA network reduce monthly prices and data allowances. Customers can add more data via “Bridge Data.” Smartphone plans start at $45/month for unlimited voice, text and 500MB of data. The plan for feature phones includes 500 voice minutes, unlimited text and 500MB of data for $35/month. Id. Customers can add 500MB for $5 with a 30-day expiration and 1GB for $10 or 3GB for $20 with a 90-day expirationlxv 3/14 Sprint Prepaid Smart New brand launched with two smartphone plans: unlimited voice, text and WiFi-enabled data only for $45/month; or unlimited voice, text and unlimited data for $60/month, with speed reductions after 2.5GB of data use. Video streaming may be limited to 3G speeds on the 2.5GB data tier.lxvi 3/14 Sprint Boost, Virgin Slower speeds for heavy data users: speeds of users who exceed 2.5GB of data usage in a monthly cycle will be reduced to 128 Kbps for the remainder of the billing cycle, down from previous 256 Kbps, unless they top up their accounts.lxvii 3/14 AT&T Cricket FCC approves AT&T’s acquisition of Leap. AT&T agreed to offer a 141 prepaid plan for feature phones that includes unlimited voice, text and data (with speeds slowed after 500 MB) for $40 per month for at least 18 months. AT&T begins process of migrating Cricket customers off Leap’s legacy CDMA network onto AT&T’s nationwide GSM-based network.lxviii 4/14 AT&T Aio Group discount plans launched that offer customers progressively increasing monthly discounts for adding new lines to an existing account. All accounts pay the full monthly price on the first line, with progressively increasing discounts applied to each new additional line: a $10/month discount on the second line, a $20/month discount on the third line and a $30/month discount on the fourth and fifth lines. The discounts are available on plans starting at $40/month. lxix . 5/14 Sprint Boost Prices cut through introduction of new ‘Monthly Unlimited Select’ plans at lower price points than legacy plans. All plans include unlimited voice and text, plus one of three data options -- 500MB for $40/month, 2.5GB for $50/month and 5GB for $60/month. Data speeds are slowed down to 2G after thresholds are reached. The new plans replace $50/month feature phone plan and $55/month smartphone plan. The ‘Bill Shrink’ plan, which was launched in March 2012, rewarded customers for making on- time payments by progressively reducing their monthly bill, up to a maximum discount of $15/month. ‘Bill shrink’ option discontinued. lxx 5/14 AT&T Aio/New Cricket Existing Aio Wireless prepaid brand shut down and re-launched as Cricket. Re-launch of Cricket brand on nationwide network begins.lxxi 5/14 AT&T New Cricket New rate plans similar to Aio’s plans introduced: 500MB for $40/month, 2.5GB for $50/month, and 5GB and international texting for $60/month. Plans include unlimited voice and text.lxxii Peak download speeds of 8 Mbps on LTE, and 4 Mbps on HSPA+, reduced to 128 Kbps if customers exceed monthly data allowance.lxxiii $5/month discount if customer selects autopay option. Monthly rates drop to $35, $45 and $55.lxxiv Group discounts: discounted monthly rates available for additional lines within a family or group of friends. As with Aio group discount plans, group discounts provide a $10/month discount for each new line of service added to a single account. lxxv Customers who make 12 on-time payments in a year offered a $50 device credit or reward that can be used to upgrade to new handset.lxxvi 5/14 AT&T New Cricket Handset financing options for prepaid phones launched. All three financing options are available for phones $149.98 and higher, with accessories excluded from eligibility.lxxvii Tier I installment loan: 0% APR for 24 months with minimum monthly payment of 4.16% of initial balance; Tier II installment loan: 29.99% APR for 6 months deferred interest and 24-month term and minimum monthly payment of 5.4% of initial balance; Lease to own: 12-month lease term, at end of which customer owns device. 90-day same-as-cash payoff option. 6/14 T-Mobile GoSmart Price cut on high-end plan – $40/month for unlimited voice, text and the 142 first 3GB of data at 3G speeds before reducing speed, down from $45/month.lxxviii Data speeds of $35 mid-level plan increased – unlimited voice, text and the first 500MB of data at 3G speeds before reducing speed to 2G speeds. Data users were previously limited to 2G speeds.lxxix 6/14 T-Mobile Brightspot Price cuts and other revisions to offerings: lxxx Unlimited voice and messaging for $30/month, down from $35/month. New data-centric plan with 300 voice minutes, unlimited messaging and 3GB of data at HSPA+ or LTE speeds before reducing speeds for $35/month. Unlimited voice, messaging and 1GB of data at HSPA+ or LTE speeds before reducing speeds for $45/month, down from $50/month. Unlimited voice, messaging and 3GB of data at HSPA+ or LTE speeds before reducing speeds for $55/month, replacing previous $65 high-end plan with 4GB of LTE data before reducing speeds. 7/15 Verizon Allset Verizon added LTE service for its Allset prepaid plans on Tuesday. Previously, Verizon’s prepaid customers were only allowed on the CDMA network. Prices are staying the same as before: the base plan, which includes unlimited calls, texts, and 500MB of mobile data, costs $45 and you get the option to add 1GB and 3GB blocks of rollover data. LTE speeds will require an LTE-capable device, and you can either bring your own or purchase one from Verizon. If you’ve got an XLTE-capable device, you can take advantage of the added speeds from Verizon’s new LTE network as well.lxxxi 143 Table V.B.ii AT&T and Cricket Prepaid Plan Offerings As of June 2014 GoPhone Monthly Smartphone Planslxxxii Price Talk Text International Text* Data $60/month Unlimited Unlimited Unlimited 2.5GB (Add more: $10/1GB) $40/month 500 minutes Unlimited Unlimited 500MB (Add more: $5/100MB) $25/month 250 minutes Unlimited Unlimited Add a data package, $5/50MB GoPhone Basic and Messaging Phone Plans Price Talk Text International Text* Data $45/month Unlimited Unlimited Unlimited Unlimited: First 2GB at speeds up to 3G $25/month 250 minutes Unlimited Unlimited Pay per use or add $5/50MB package $2/day, only on days used Unlimited Unlimited Unlimited Pay per use 1¢/5KB 10¢/minute 10¢/minute 20¢/text or buy package 25¢/text message Pay per use 1¢/5KB Cricket Smartphone and Basic Cell Phone Planslxxxiii Price Talk Text International Text* Data498 Pro: $60/month* Unlimited Unlimited Unlimited Texting to 35 countries 5GB at high-speed** Smart: $50/month* Unlimited Unlimited 2.5 GB at high-speed** Basic: $40/month* Unlimited Unlimited 500MB at high-speed ** Talk & Text: $25/month Unlimited Unlimited Available if you add Cricket Passport ($15/ month, text to 35 countries) Basic phones only - does not include data or multimedia messaging #Eligibility and Monthly Discount on Cricket plans: Customers in good standing with at least two (2) eligible lines of service on one account will be automatically enrolled to receive Group Save discounts. Only eligible lines two-five on the same account receive the monthly discount described below. Only lines of service on a Basic, Smart, or Pro rate plan are eligible. Your first eligible line does not receive a discount; the second eligible line = $10 monthly discount; the third eligible line = $20 monthly discount; and the fourth and fifth eligible lines each = $30 monthly discount. The maximum monthly discount is $90 when you have five (5) eligible lines. *$5 Auto-pay credit. Eligible for group save discounts.MMS not included ***Unlimited data available at reduced speeds after high-speed data limit is reached 498 Add 1 GB high-speed for $10/month 144 Table V.B.iii Verizon Wireless Prepaid Plans As of June 2014 Verizon Allset Smart Phone Monthly Prepaid Plan Price Talk Text Data* $45/month Unlimited Unlimited 500 MB Verizon Allset Basic Phone Monthly Prepaid plan Price Talk Text Data* $45/month Unlimited Unlimited 500 MB $35/month 500 minutes Unlimited 500 MB *Data does not include 4G/LTE. Additional data may be purchased: 500 MB for $5 (expires after 30 days); 1 GB for $10 (expires after 90 days); or 3 GB for $20 (expires after 90 days) Verizon Wireless Pay-as-you-go Plans Price Mobile to Mobile Calling (with Verizon Wireless customers) Talk Text/Picture Messaging (price per minute) Mobile Web* $1.99/day Unlimited Unlimited 2¢/25¢ 99¢/day 99¢/day Unlimited 10¢/minute 10¢/25¢ 99¢/day pay as you go 25¢/minute 25¢/minute 20¢/25¢ 99¢/day Table V.B.iv Sprint Prepaid Planslxxxiv As of June 2014 Sprint Prepaid Monthly Smartphone Plans* Plan/Price Talk Text High speed 3G/4G data** Smart Plus: $60/month Unlimited Unlimited 2.5GB/month of high-speed data. Smart: $45/month Unlimited Unlimited Data not included Sprint Prepaid Monthly Basic Plans* Plan/Price Talk Text Data Basic Plus ($50/month) Unlimited Unlimited Unlimited Basic ($35/month) 500 minutes Unlimited Unlimited *All Prepaid Plans include voicemail, long distance, call waiting, call forwarding ** When monthly usage exceeds applicable data plan allotment, speeds (including video) reduced to 2G speeds for remainder of plan cycle Boost Mobile Monthly Prepaid Plans Price Talk Text High speed 3G/4G data* $60/month Unlimited Unlimited 5 GB. . $50/month Unlimited Unlimited 2.5 GB $40/month Unlimited Unlimited 500 MB Boost Mobile Daily Prepaid Plans Price Talk Text Data $ 3/day (smartphones ) Unlimited Unlimited Unlimited $2/day (basic phones) Unlimited Unlimited Unlimited *All Prepaid Plans include voicemail, long distance, call waiting, call forwarding * When monthly usage exceeds applicable data plan allotment, speeds (including video) reduced to 2G 145 speeds for remainder of plan cycle Boost Shrinking Payment Plan Price* Talk--** Text Data*** $60 initially, falls to $45 Unlimited Unlimited (includes BBM Voice) Unlimited $55 initially, falls to $40 Unlimited Unlimited Unlimited $50 initially, falls to $35 Unlimited Unlimited Unlimited $45 initially, falls to $30 Unlimited Unlimited Unlimited *Shrinking Payments reduces your payment by $5/month for every six on-time payments. **Plans include voicemail, long distance, call waiting, call forwarding ***Includes 2.5GB/month of high-speed data. Adaptive Protocol Video limited to 3G Virgin Mobile Beyond Talk Unlimited Plans (Android Compatible) Price Talk Text Data (4G data where available) $55/month Unlimited Unlimited Unlimited $45/month 1200 minutes Unlimited Unlimited $35/month 300 minutes Unlimited Unlimited Virgin Mobile PayLo Talk & Text Plans (Not Android Compatible) Plan & Price Talk Text Data $40/month Unlimited Unlimited 50 MB included $30/month 1500 minutes 1500 messages 30 MB included $20/month 400 minutes 15¢/message $1.50/MB Table V.B.v Selected T-Mobile and Metro PCS Prepaid Plan Offeringslxxxv As of June 2014 T-Mobile Monthly and Daily Prepaid Planslxxxvi Price Talk Text Data (4G LTE*) $80/month Unlimited Unlimited Unlimited $70/month Unlimited Unlimited 5 GB $60/month Unlimited Unlimited 3 GB $50/month Unlimited Unlimited 1 GB $40/month* Unlimited Unlimited 500 MB $30/month 100 minutes Unlimited 5 GB $35/month Unlimited Unlimited NA $3/day Unlimited Unlimited 200 MB $2/day Unlimited Unlimited Only 2G available *Simple starter plan MetroPCS Monthly Prepaid Plans#lxxxvii Price* Talk Text Data (4G LTE**) $60/month Unlimited Unlimited Unlimited $50/month Unlimited Unlimited 2.5 GB $40/month Unlimited Unlimited 500 MB 146 *Save $5 per line/month with a family plan, a family can consist of up to 5 people (a total savings of $25/month) **Data are still available at reduced speeds after reaching the data limit # MetroPCS offers pay by minute and pay by week plans, only in brick and mortar stores. GoSmart Monthly Prepaid Planslxxxviii Price Talk* Text** Data Facebook and Facebook Messenger $45/month Unlimited Unlimited 3GB at 3G speeds, then 2G speeds for the rest of the cycle Unlimited $35/month Unlimited Unlimited 500 MB at 3G speeds, then 2G speeds for the rest of the cycle Unlimited $25/month Unlimited Unlimited NA Unlimited T-Mobile Pay-as-you-golxxxixwithout Data Price Talk Text Minutes Cost/min Duration $100 1000 10¢/min 1 year 10¢ to send/receive texts, 25¢ to send/receive pictures & videos $50 400 13¢/min 90 days $30 160 19¢/min 90 days $10 30 33¢/min 90 days * Optional $10/month international texting and calling bundle ** Optional $5/month international texting Table V.B.vi US Cellular Prepaid Plansxc July 2014 US Cellular Simple Connect Smart Phone Monthly Prepaid Plan Price Talk Text Data $60/month Unlimited Unlimited Unlimited with 2 GB at high speed, speeds reduced thereafter $50/month Unlimited Unlimited Unlimited with 500 MB at high speed, speeds reduced thereafter US Simple Connect Cellular Basic Phone Monthly Prepaid plan Price Talk Text Data $40/month Unlimited Unlimited Unlimited data for basic phones 147 Table V.B.vii C-Spire Prepaid Plansxci499 July 2014 C-Spire Pay-as-you-go Prepaid Plan Price Talk Text & Pictures Data $65/30 days Unlimited Unlimited Unlimited $55/30 days Unlimited Unlimited 1 GB $50/30 days 500 nationwide minutes Unlimited 500 MB $35/30 days 350 nationwide minutes Unlimited 500 MB C-Spire Pay-as-you-go Prepaid Plan without Data Price Talk Text & Pictures Data $50/ 30 days Unlimited Unlimited $25/ 30 days 250 nationwide minutes Unlimited Purchase data day pass for $1/day for basic phone $2/day Unlimited Unlimited 499 As of 7/7/2014, http://www.cspire.com/shop_and_learn/plans/category_plan_landing.jsp?id=cat320003#prepaid. 148 Price Indicators for Mobile Data Table V.C.i Current Postpaid Smartphone Data Pricing for Mobile Wireless Providers500 1st Quarter 2014 AT&T Verizon Sprint T-Mobile US Cellular C-Spire Base Fee $40 $40 $50 $50501 $40 $50 Data Allowance $/Mth ?/MB $/Mth ?/MB $/Mth subsidized device, 2- year contract $/Mth unsub sidize d device ?/MB $/Mth ?/MB $/Mth ?/MB $/Mth ?/MB MB 250 $15 6.0? 300 $20 6.7? $15 5.0? 500 $30 6.0? $10 2.0? GB 1 $40 4.0? $20 2.0? $40 4.0? $15 1.5? 2 $40 2.0? $50 2.5? $10 $50 2.0? 3 $60 2.0? $10 3.0? 0.5? $60 2.0? $45 1.5? 4 $70 1.8? $70 1.75? 5 $20 0.5?? $75 1.5? 6 $80 1.3? $80 1.33? 8 $90 1.13? 10 $100 1.0? $100 1.0? 12 $110 0.92? 14 $120 0.86? 15 $130 0.87? 16 $130 0.81? 500 Derived from BoA/ML 4 April 2014, “A Frantic Start to 2014 in Wireless Pricing”, Table 3, pp. 7. Numbers in the table are BofA Merrill Lynch Global Research estimates and company websites visited on 7/7/2014. 501 Basic unlimited talk and text smartphone individual plan includes1GB of data.cost $50. 3 GB plan costs $60 5 GB plan cists $70. The incremental data cost for 3 GB plan is an extra $10 for 2 GB of data, and for the 5GB plan the incremental cost is $20 for 4 GB of data as 1 Gb is already included in the base price. 149 18 $140 0.78? 20 $150 0.75? $150 0.75? 30 $225 0.75? $225 0.75? 40 $300 0.08? $300 0.75? 50 $375 0.08? $375 0.75? Source: BoA/ML, “A Frantic Start to 2014 in Wireless Pricing”, 4 April 2014 150 APPENDIX VI NON-PRICE RIVALRY Investment Table VI.A.i Quarterly Capital Expenditure by Mobile Wireless Providers (In millions) 1Q2011 – 2Q2014 1Q11 2Q11 3Q11 4Q11 1Q12 2Q12 3Q12 4Q12 1Q13 2Q13 3Q13 4Q13 1Q14 2Q14 National Operators Verizon Wireless 2,735 2,667 1,784 1,787 1,885 2,048 2,133 2,791 1,992 2,278 2,450 2,705 2,554 2,771 AT&T 1,870 2,511 2,520 2,863 2,324 2,345 2,709 3,417 2,296 3,033 3,060 2,802 3,082 3,480 Sprint 449 546 647 774 710 1,012 1,376 1,786 1,706 1,728 1,683 1,716 1,057 1,416 T-Mobile 749 688 741 551 747 539 717 898 1,076 1,050 1,017 882 947 940 Total National 5,803 6,412 5,692 5,975 5,666 5,944 6,935 8,892 7,070 8,089 8,210 8,105 7,640 8,607 Regional Operators Cincinnati Bell 5 1 5 7 6 2 4 3 8 2 2 4 6 0.6 Leap 93 93 103 152 146 119 106 63 26 22 90 130 Metro PCS 187 265 248 190 144 182 262 258 NTELOS 8 13 14 23 13 19 22 19 17 27 21 16 14 32 US Cellular 96 162 248 276 201 183 199 253 118 169 243 208 90 144 Total Regional 388 534 618 649 511 505 593 595 170 220 356 358 110 177 Total 6,191 6,946 6,310 6,624 6,177 6,449 7,528 9,487 7,240 8,309 8,566 8,462 7,750 8,784 Notes: Based on UBS Wireless 411 Report, Version 51 at 28. Metro PCS data are not available separately after the fourth quarter of 2012 as the T-Mobile and MetroPCS merger was consummated in early 2013. Leap is reported separately from AT&T as the AT&T and Leap merger was not consummated by the fourth quarter of 2013. 151 Network Coverage and Technology Upgrades 219. Appendix Tables VI.B.i – VI.B.vi contain detailed data on the percentage of the U.S. population, land area, and road miles covered by each of the top four mobile wireless service providers, and top four mobile broadband providers, and also show rural and non-rural breakdown for these categories. All tables are derived from Commission estimates based on census block analysis of Mosaik Coverage Right coverage maps, January 2014. Excludes coverage for Island Areas (Guam, American Samoa, US Virgin Islands, Mariana Islands. Population data are from the 2010 Census. The data underlying these estimates measure mobile network “coverage,” and not the number of network providers affirmatively offering service to any or all residents in those locations. Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. Table VI.B.i Estimated Mobile Wireless Coverage in the U.S. by Provider January 2014 Provider Number of Blocks POPS Contained in Those Blocks % of Total US POPs % of Total US Square Miles % of Total US Road Miles ATT 10,324,435 308,396,818 98.7 61.8 85.3 VZW 10,164,336 302,838,131 96.9 61.2 84.8 Sprint 7,606,057 280,488,603 89.8 24.9 48.1 T-Mobile 8,286,265 287,702,030 92.1 34.3 58.3 Source: Mosaik January 2014, Census 2010.The data underlying these estimates measure mobile network “coverage,” and not the number of network providers affirmatively offering service to any or all residents in those locations. Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. Table VI.B.ii Estimated Mobile Wireless Network Coverage, Selected Facilities-Based Providers Voice Networks, 2009-2014 (Covered POPs, in millions) Service Provider Oct. 2009 Oct. 2010 Apr. 2011 Jan. 2012 Oct. 2012 Jan. 2014 AT&T 262.8 281.9 306.3 306.6 307.2 306.9 Verizon Wireless 270.5 284.9 299.5 299.5 300.0 300.6 T-Mobile 246.2 249.5 282.5 284.8 281.4 286.7 Sprint Nextel 258.0 263.2 292.1 291.2 290.3 275.2 MetroPCS 84.6 92.1 105.0 105.4 108.1 - Leap 80.5 82.7 94.0 93.4 94.2 - US Cellular 41.7 41.5 44.2 44.0 44.0 314.6 Note: The estimates in this Table are based on our census block analysis of Mosaik CoverageRight coverage maps using the April 2011, January 2012, and October 2012 data. The population data are from the 2010 Census. Estimates for 2009 and 2010 are obtained from the Fifteenth Report, 25 FCC Rcd at 9702 ¶ 45. The data underlying these estimates measure mobile network “coverage,” and not the number of network providers affirmatively offering service to any or all residents in those locations. Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. 152 Table VI.B.iii Estimated Mobile Wireless Broadband Coverage in the U.S. by Provider January 2914 Provider Number of Blocks POPS Contained in Those Blocks % of Total US POPs % of Total US Square Miles % of Total US Road Miles ATT 9,920,357 305,640,460 97.8 55.1 79.2 VZW 10,113,892 302,645,159 96.9 59.5 83.6 Sprint 7,507,299 278,989,810 89.3 24.5 47.4 T-Mobile 5,603,484 246,362,777 78.8 15.0 32.3 . Source: Mosaik January 2014, Census 2010. The data underlying these estimates measure mobile network “coverage,” and not the number of network providers affirmatively offering service to any or all residents in those locations. Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. Table VI.B.iv Estimated Mobile Wireless Network Coverage, Selected Facilities-Based Providers Broadband Networks, 2009-2014 (Covered POPs, in millions) Service Provider Nov. 2009 Aug. 2010 Apr. 2011 Jan. 2012 Oct. 2012 Jan. 2014 AT&T 212.3 228.6 276.1 289.9 296.7 305.7 Verizon Wireless 266.7 270.0 298.0 299.2 300.4 302.6 Sprint Nextel 226.9 239.4 276.4 273.7 275.1 279.1 T-Mobile 133.9 183.8 214.7 227.6 235.4 246.3 MetroPCS - - 62.2 72.4 108.3 - Clearwire - - 108.9 105.1 105.3 - Leap 79.2 81.5 92.6 92.3 93.4 - US Cellular 26.6 30.0 40.7 41.1 43.2 311.9 Note: For purposes of this, and earlier, Mobile Wireless Competition Reports, we include coverage by WCDMA/HSPA, HSPA+, EV-DO, WiMAX, and LTE networks within our estimate of mobile broadband network coverage. Commission estimates based on census block analysis of Mosaik CoverageRight coverage maps, April 2011, January 2012, and October 2012. Population data are from the 2010 Census. Estimates for 2009 and 2010 are obtained from the Fifteenth Report. Fifteenth Report, 25 FCC Rcd at 9702 ¶ 45. The recent Broadband Progress Report did not include WCDMA/HSPA or EV- DO networks in its definition of mobile broadband networks. 2012 Eighth Broadband Progress Report, WN Docket No. 11- 121 (rel. Aug. 21, 2012) ¶ 40. The data underlying these estimates measure mobile network “coverage,” and not the number of network providers affirmatively offering service to any or all residents in those locations. Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. 153 Table VI.B.v Estimated Mobile Wireless Voice Coverage in Rural Areas by Provider January 2014 Number of Rural Census Blocks POPS Contained in Rural Census Blocks % of Total U.S. POPs % of Total U.S. Square Miles % of Total U.S. Road Miles Total for US 5,387,335 59,151,859 18.9 84.5 67.3 Provider Number of Rural Census Blocks POPS Contained in Rural Census Blocks % of Total Rural U.S. POPs % of Total Rural U.S. Square Miles % of Total Rural U.S. Road Miles ATT 4,349,508 53,931,208 91.2 49.3 72.1 VZW 4,404,874 52,710,410 89.1 54.1 76.1 Sprint 2,153,983 34,185,221 57.8 15.3 27.7 T-Mobile 2,763,501 38,999,396 65.9 25.3 41.1 Source: Mosaik January 2014, Census 2010. The data underlying these estimates measure mobile network “coverage,” and not the number of network providers affirmatively offering service to any or all residents in those locations. Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. Table VI.B.vi Estimated Mobile Wireless Voice Coverage in Non-Rural Areas by Provider January 2014 Number of Non-Rural Census Blocks POPS Contained in Non-Rural Census Blocks % of Total U.S. POPs % of Total U.S. Square Miles % of Total U.S. Road Miles Total for US 5,768,151 253,319,468 81.8 15.5 32.7 Provider Number of Non-Rural Census Blocks POPS Contained in Non-Rural Census Blocks % of Total Non-Rural U.S. POPs % of Total Non-Rural U.S. Square Miles % of Total Non-Rural U.S. Road Miles ATT 5,716,694 252,900,141 99.8 89.4 97.7 VZW 5,582,339 247,913,734 97.9 85.7 95.3 Sprint 5,158,015 240,994,632 95.1 60.6 79.8 T-Mobile 5,425,720 247,797,133 97.8 74.8 89.4 Source: Mosaik January 2014, Census 2010. The data underlying these estimates measure mobile network “coverage,” and not the number of network providers affirmatively offering service to any or all residents in those locations. Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. . 154 Table VI.B.vii Estimated Mobile Broadband Coverage in Rural Areas by Provider January 2014 Number of Rural Census Blocks POPS Contained in Rural Census Blocks % of Total U.S. POPs % of Total U.S. Square Miles % of Total U.S. Road Miles Total for US 5,387,335 59,151,859 18.9 84.5 67.3 Provider Number of Rural Census Blocks POPS Contained in Rural Census Blocks % of Total Rural U.S. POPs % of Total Rural U.S. Square Miles % of Total Rural U.S. Road Miles ATT 4,235,577 53,085,491 89.7 49.3 70.8 Verizon 4,503,652 53,936,545 91.2 54.6 77.5 Sprint 2,260,761 35,283,398 59.6 17.0 30.1 T-Mobile 991,184 17,388,659 29.4 8.4 14.1 Source: Mosaik January 2014, Census 2010. The data underlying these estimates measure mobile network “coverage,” and not the number of network providers affirmatively offering service to any or all residents in those locations. Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. Table VI.B.viii Estimated Mobile Broadband Coverage in Non-Rural Areas by Provider January 2014 Number of Non-Rural Census Blocks POPS Contained in Non-Rural Census Blocks % of Total U.S. POPs % of Total U.S. Square Miles % of Total U.S. Road Miles Total for US 5,768,151 253,319,468 81.8 15.5 32.7 Provider Number of Non-Rural Census Blocks POPS Contained in Non-Rural Census Blocks % of Total Non-Rural U.S. POPs % of Total Non-Rural U.S. Square Miles % of Total Non-Rural U.S. Road Miles ATT 5,684,780 252,554,969 99.7 86.3 96.4 Verizon 5,610,240 248,708,614 98.2 86.6 96.0 Sprint 5,246,538 243,706,412 96.2 65.5 83.0 T-Mobile 4,612,300 228,974,118 90.4 50.8 69.7 Source: Mosaik January 2014, Census 2010. The data underlying these estimates measure mobile network “coverage,” and not the number of network providers affirmatively offering service to any or all residents in those locations. Coverage calculations based on Mosaik data, while useful for measuring developments in mobile wireless coverage, have certain limitations that likely overstate the extent of mobile wireless coverage. 155 C. Quality of Service i. Ookla 1. Description of Ookla Speed Test. The Ookla speed test is solely crowdsourced and requires users to choose to run each individual test. This app is available free of charge for iOS, Amazon, Android, and Windows Phones. It tests latency, download speed and upload speed. Ookla Net Index mobile app users are able to perform speed tests any time they have a wireless connection. 2. To measure download speed, the tested device first downloads small binary files from the web server to the client, and Ookla measures that download to estimate the connection speed. There are several factors that can affect the speed of individual samples, therefore the fastest 10 percent and slowest 10 percent of the samples are discarded. In addition, because the ramp-up period can take a significant part of the beginning of the test, an additional 20 percent of the bottom samples are trimmed. Overall, the fastest 10 percent and slowest 30 percent of the samples are discarded. The remaining samples are averaged together to determine the final reported result.502 To measure upload speed, a small amount of random data is first generated in the client device and sent to the web server to estimate the connection speed. Based on this result, an appropriately sized chunk of randomly generated data are selected for upload. The upload test is then performed in chunks of uniform size. The slowest 50 percent of the observations are dropped, and the fastest 50 percent of the observations are averaged to eliminate anomalies and determine the result. 503 3. The latency test simply sends HTTP requests to the selected server, and measures the time it takes to get a response.504 The Ookla speed test chooses a server for each test individually, based on which server from a set of possible test servers has the fastest latency. However, the tester also has the ability to choose the server that the test will use. 4. Presentation of Ookla Speed Test Data. For this presentation, we use the city-level Net Index speed data, which presents daily mean upload and download speeds by city.505 The daily, city-level observations can consist of thousands of speed measurements, averaged into one data point.506 Because this dataset is aggregated at the city level on a daily basis, we cannot identify individual connection speeds. 5. In this Report, we present mobile wireless upload and download speeds within the United States for 2013, and for the first half of 2014. We estimate nationwide speeds by service provider. We also estimate California-only speeds by service provider, in order to facilitate comparison with CalSPEEDdata. Not all wireless providers are represented in the Net Index data. For example, there were no observations for US Cellular during this time frame.507 Similarly, not all states are represented during these years, due to the lack of Ookla mobile app users in certain states. For instance, Alaska, North Dakota, West Virginia, Wyoming, and Vermont are not represented in the 2014 data. 6. We calculate the median508 and mean upload and download speeds by service provider. The 502 From: https://support.speedtest.net/entries/20862782-How-does-the-test-itself-work-How-is-the-result-calculated- 503 From: https://support.speedtest.net/entries/20862782-How-does-the-test-itself-work-How-is-the-result-calculated- 504 From: https://support.speedtest.net/entries/20862782-How-does-the-test-itself-work-How-is-the-result-calculated- 505 Net Index speed data are available at the city level (city_isp_daily_speeds.csv), the state level (region_isp_daily_speeds.csv), and the country level (country_isp_daily_speeds.csv) at http://www.netindex.com/#source ; http://www.ookla.com/support/a39030078/Frequently-Asked-Questions 506 The free dataset that is publicly available from Ookla consists of anonymous daily index values. The dataset reports the average speed, the number of tests that generate the speed and the date when the tests were done for every geographic location found at NetIndex.com. See http://netindex.com/source-data/. There is also available for purchase a more extensive dataset from Ookla that contains every individual speed test measurement. 507 MetroPCS was dropped from this analysis, as there were only 10 observations for this provider from 2012 until the present. We also drop a few cities due to insufficient observations. 508 The median speed is actually a median of daily averages, and therefore does not represent a true median speed. 156 estimated nationwide download speeds by service provider are presented in table VI.C.i, and the estimated nationwide upload speeds by service provider are presented in table VI.C.ii. The estimated California-only upload speeds by service provider are presented in table VI.C.iii, and the estimated California-only download speeds by service provider are presented in table VI.C.iv. Table VI.C.i Ookla - Estimated Download Speeds by Service Provider, Nationwide Service Provider 2013 Jan - June 2014 Mean down load speed (Mbps) Median down load speed (Mbps) Number of tests (’000s) Mean down load speed (Mbps) Median down load speed (Mbps) Mumber of tests (’000s) Verizon 9.13 8.99 30,200 10.81 10.74 12,500 AT&T 9.21 9.04 7,588 9.58 9.79 2,955 Sprint 2.26 1.92 28,800 3.2 2.9 11,700 T-Mobile 6.48 6.16 4,719 9.96 9.89 3,146 Source: Ookla NetIndex data Table VI.C.ii Ookla - Estimated Upload Speeds by Service Provider, Nationwide Service Provider 2013 Jan - June 2014 Mean upload speed (Mbps) Median upload speed (Mbps) Number of tests (’000s) Mean upload speed (Mbps) Median upload speed (Mbps) Number of tests (’000s) Verizon 4.22 4.14 30,200 5.41 5.37 12,500 AT&T 3.97 3.86 7,588 4.44 4.55 2,955 Sprint 0.83 0.67 28,800 1.09 0.89 11,700 T-Mobile 2.03 1.76 4,719 4.22 4.16 3,146 Source: Ookla NetIndex data Table VI.C.iii Ookla - Estimated Download Speeds by Service Provider, California Only. Service Provider 2013 Jan - June 2014 Mean down load speed (Mbps) Median down load speed (Mbps) Number of tests (‘000s) Mean down load speed (Mbps) Median down load speed (Mbps) Number of tests (‘000s) Verizon 8.51 8.49 6,973 9.82 9.82 2,822 AT&T 9.56 9.27 1,230 9.71 9.42 6445 Sprint 1.77 1.22 6,888 2.94 2.38 2,069 T-Mobile 7.39 7.03 944 10.49 10.35 886 Source: Ookla NetIndex data 157 Table VI.C.iv Ookla - Estimated Upload Speeds by Top 4 Nationwide Service Provider, California only. Service Provider 2013 Jan - June 2014 Mean upload speed (Mbps) Median upload speed (Mbps) Number of tests (‘000s) Mean upload speed (Mbps) Median upload speed (Mbps) Number of tests (‘000s) Verizon 4.47 4.45 6,973 5.60 5.56 2,822 AT&T 4.17 4.04 1,230 4.82 4.58 645 Sprint 0.70 0.55 6,888 1.02 0.67 2,069 T-Mobile 2.54 2.33 944 4.66 4.63 886 Source: Ookla NetIndex data ii. FCC 7. The FCC Speed Test app is available free of charge for Android phones and for the iPhone. The application measures mobile broadband performance in four active categories: download speed, upload speed, latency and packet loss. Several other passive metrics are also recorded, including signal strength of the connection, and device manufacturer and model.509 The FCC speed test can be set to automatically run in the background of Android phones, but not iPhone devices, for which the user must execute the speed test manually. No tests are executed if the device is transferring more than 64kbit/s at the time a test is scheduled to execute. Tests that are skipped are rescheduled to execute at a later time. By default the app will limit the total monthly data traffic used for execution of scheduled tests to a maximum of 100MB. Volunteers can adjust the data cap to suit their preference.510 8. When starting a measurement cycle, the application runs a brief latency test to measurement servers in the application's configuration. The nearest measurement server with the lowest round-trip latency is selected as the target for all subsequent measurements (throughput, latency and packet loss).511 In a manner similar to the Ookla test, the FCC test makes an adjustment to account for possible slow TCP startup. If a packet is not received back within three seconds of sending, it is treated as lost.512 9. The speed test measures the download and upload speed of the given connection in bits per second by performing multi-connection GET and POST HTTP requests to a target test node. Binary non-zero content, herein referred to as the payload, is hosted on a web server on the target test node. The test operates for either a fixed duration (in seconds) or a fixed volume (in MB). It can also report the recorded average throughput at multiple intervals during the test (e.g. once every five seconds). The client will attempt to download as much of the payload as possible for the duration of the test.513 The speed test preconditions streams to get past possible slow TCP startup, which can bias the throughput test towards reporting higher throughput.514 Further details regarding the methodology used in presenting FCC Speed Test app data are provided in the text of this Report. iii. RootMetrics 10. RootMetrics tests data, call, and text performance in all 50 states across the United States. As described by the company, its testing of data performance measures reliability and speed for file uploads and 509 http://www.fcc.gov/measuring-broadband-america/mobile 510 http://www.fcc.gov/measuring-broadband-america/mobile 511 http://www.fcc.gov/measuring-broadband-america/mobile/technical-summary 512 http://www.fcc.gov/measuring-broadband-america/mobile/technical-summary 513 http://www.fcc.gov/measuring-broadband-america/mobile/technical-summary 514 CalSPEED: Measuring California Mobile Broadband - A Comparison of Methodologies 158 downloads, for downloading email, and for loading lite data files that represent loading typical webpages and apps.515 Its call testing measures how reliably each network is able to place and maintain calls.516 Finally, its text testing measures how reliably and quickly consumers can send and receive text messages.517 RootMetrics uses these measurements to calculate a Reliability Index and a Speed Index, denoted as RootScores, using a proprietary algorithm that it describes as focusing on large-scale patterns of performance and minimizing the effects of isolated, rare events.518 These RootScores are available for each of the airports and metropolitan areas included in the sample, as well as at the state and national levels.519 Finally, RootMetrics combines these RootScores using a proprietary method to compute an overall performance score. In this section we discuss only the national Speed Index data for the top four facilities-based providers. 11. The RootMetric Speed Index takes into account the speed measurements of both data and texts. The data speed measurements assess how quickly consumers can connect to the network, as well as how quickly data tasks can be completed online (downloading/uploading files, downloading email, and downloading lite data files that approximate loading a typical webpage or app). The text speed measurements assess how quickly consumers can send and receive text messages.520 The raw data speed measurements and text speed measurements are then converted by RootMetrics into separate speed indices for data and for text. These separate indices are then combined and converted into overall speed indices using a proprietary algorithm. According to RootMetrics, RootScores (i.e. the indices) are meant to reflect a consumer’s experience of network performance and are scaled from 0 – 100, with the lower limit representing network performance that would result in a poor consumer experience and the upper limit reflecting extraordinary performance.521 Table V.C.v provides the national Speed Index data for the four nationwide providers and their separate data and text components for the second half of 2013 and the first half of 2014.522 Table V.C.vi provides the same data for California. 515 During file transfer testing, RootMetrics attempts to open an HTTP connection(s), and then measures network connection success rates, as well as upload and download transfer speeds. The testing measures how reliably and quickly each network is able to: 1) connect to an IMAP server and download a group of 10 emails and 2) establish a network connection and download lite data files to represent typical web and app behaviors. http://www.rootmetrics.com/us/methodology 516 To measure call performance, RootMetrics places a call from each network’s phone and attempts to hold that call open for the duration of the test cycle. The testing shows blocked and dropped outgoing call failure rates. http://www.rootmetrics.com/us/methodology 517 To analyze texting, RootMetrics measures send failure rates and the speed at which each network can send and receive texts from a phone within its own network and phones within the other networks. http://www.rootmetrics.com/us/methodology 518 http://www.rootmetrics.com/us/methodology 519 http://www.rootmetrics.com/us/methodology 520 http://www.rootmetrics.com/us/methodology 521 http://www.rootmetrics.com/us/standards. Prior to January 2014, Data RootScores in Metro and Airport RootScore Reports could exceed 100 if performance was extraordinary, http://www.rootmetrics.com/us/methodology. 522 Source: http://www.rootmetrics.com/us/rsr/united-states/2013/2H and http://www.rootmetrics.com/us/rsr/united- states/2014/1H. 159 Table VI.C.v RootMetrics: National Speed Index Data 2nd Half 2013, 1st Half 2014 2nd Half 2013 1st Half 2014 Speed Index Data Performance Text Performance Speed Index Data Performance Text Performance AT&T 88.7 83.9 94.3 71.1 78.1 81.2 Verizon 88.5 87.3 94.7 75.7 81.5 80.4 Sprint 65.8 51.2 92.8 54.2 63.7 78.8 T-Mobile 74.3 48.7 90.1 64.4 67.1 78.4 Source: RootMetrics RootScore Report Data, 2nd half 2013, 1st half 2014. Table VI.C.vi RootMetrics: California Speed Index Data 2nd Half 2013, 1st Half 2014 2nd Half 2013 1st Half 2014 Speed Index Data Performance Text Performance Speed Index Data Performance Text Performance AT&T 85.3 80.4 92.2 83.0 89.6 87.9 Verizon 88.5 84.3 92.8 88.9 93.7 88.6 Sprint 62.6 44.7 89.3 62.3 72.0 87.2 T-Mobile 82.5 65.9 90.2 83.0 86.3 87.3 Source: RootMetrics RootScore Report Data, 2nd half 2013, 1st half 2014. iv. CalSPEED 12. Description of CalSPEED. CalSPEED is a structured sampling program of 1,986 locations (originally 1,200) scattered throughout California. The sites are spread across urban (37 percent), rural (56 percent) and tribal (seven percent), lands. These sites are visited every six months and speed tests are run on the latest available Android phone and also on a USB network device on a Windows based netbook, for each of the four major providers (Verizon, AT&T, Sprint, and T-Mobile).523 Tests are not performed using iPhones. CalSPEED data was collected in five rounds from the spring of 2012 through Spring 2014.524 13. Each test in the structured sampling program is run using the same protocol. Tests are performed inside a stationary automobile. First, a valid GPS reading is obtained from the GPS receiver connected to a netbook, and tests are run using the data cards for each provider. Next, the smart phone tests are performed. Results are uploaded to the cloud-based database server at each location. In cases where data cannot be uploaded from the tested location, the test results remained on the netbook or smart phone until the tester reaches a location with sufficient network coverage to upload the data.525 14. Although not yet incorporated into CalSPEED’s published speed test results, in April 2013 CalSPEED launched a free mobile speed test app that is available for download on Android phones.526 The 523 From CalSPEED: California Mobile Broadband - An Assessment. Ken Biba Managing Director and CTO Novarum, Inc. 524 Spring 2013 tests were taken between the dates of 4/4/2013 to 4/29/2013, while Fall 2013 tests were taken between the dates of 10/17/2013 to 12/18/2013. Spring 2014 tests were taken between the dates of 4/10/2014 and 6/05/2014. 525 California Public Utilities Commission: Spring 2012 Mobile Broadband Field Testing Initial Staff Report. September, 2012. pp. 3. 526 http://calbroadbanddrivetest.blogspot.com/2013/04/calspeed-is-now-available-on-google-play.html 160 smartphone user must manually run this test, as it does not automatically run in the background of the phone. The CalSPEED crowdsourced data are not included in our analysis.527 15. CalSPEED measures the complete network path, from the client device, through the local access network, through the Internet backbone, to two fixed servers. One server is physically located in Northern California and the other in Northern Virginia - both in the Amazon Web Services cloud.528 CalSPEED measures network metrics including end-to-end packet latency, upload speed, download speed, packet loss and jitter. These data track three major trends over time: changes in performance (throughput, latency and jitter) due to new technology and capacity deployment, changes in performance due to increases in user load, and changes in coverage as providers deploy their footprint. Observations are not included in the analysis if the measurement was taken outside of the carrier’s coverage area, or if the tester did not complete the test. Any other errors are counted as zero throughput. CalSPEED reduces their calculated means by one standard deviation. 16. In this Report, we present mobile wireless upload and download speeds and latency within the United States for 2013 through Spring 2014. The estimated download speeds by provider are presented in Chart VI.C.i, and the estimated upload speeds by provider are presented in Chart VI.C.ii. In terms of both median and mean speeds, Verizon has the fastest download and upload speeds in 2013, followed by AT&T, T-Mobile, and then Sprint. For the period spanning fall 2013 and spring 2014, Verizon still has the highest mean and median download and upload speeds. T-Mobile shows significant speed improvements in Spring 2014, surpassing AT&T in both mean and median download speeds. Chart VI.C.i CalSPEED Mean Downstream Throughput, 2012-2014 Source: “CALSPEED - California’s Mobile Broadband Assessment”, November 2014. Chart VI.C.ii CalSPEED Mean Upstream Throughput, California, 2012-2014 527 From CalSPEED: California Mobile Broadband - An Assessment. Ken Biba Managing Director and CTO Novarum, Inc. 528 From CalSPEED: California Mobile Broadband - An Assessment. Ken Biba Managing Director and CTO Novarum, Inc. 161 Source: “CALSPEED - California’s Mobile Broadband Assessment”, November 2014. Chart VI.C.iii CalSPEED Mean Latency, California, 2012-2014 Source: “CALSpeed - California’s Mobile Broadband Assessment”, November 2014. APPENDIX VII LIST OF ABBREVIATIONS AND ACRONYMS 2G Second Generation 3G Third Generation 4G Fourth Generation ALMB Average Local Monthly Bill 162 ARPU Average Revenue Per User ATN Atlantic Tele-Network AWS Advanced Wireless Service BEA Bureau of Economic Analysis BLS Bureau of Labor Statistics BRS Broadband Radio Service BTA Basic Trading Area CFR Code of Federal Regulations CAPEX Capital Expenditures CDC Centers for Disease Control CDMA Code Division Multiple Access CEA Component Economic Area CEO Chief Executive Officer CMA Cellular Market Area CMRS Commercial Mobile Radio Services CPI Consumer Price Index DA Delegated Authority DAS Distributed Antenna System DOJ Department of Justice DSL Digital Subscriber Line EA Economics Area EBIT Earnings before Interest and Taxes EBITDA Earnings before Interest, Taxes, Debt, and Amortization EBS Educational Broadband Service EDGE Enhanced Data Rates for Global Evolution EHA Exclusive Handset Agreement ETF Early Termination Fee EV-DO Evolution Data Optimized FCC Federal Communications Commission FDD Frequency Division Duplex FNPRM Further Notice of Proposed Rulemaking FSS Frequency Spread Spectrum FTC Federal Trade Commission GAO Government Accountability Office GB Gigabyte GHz Gigahertz GPRS General Packet Radio Service GSM Global System for Mobile Communication HHI Herfindahl-Hirschman Index HSPA High Speed Packet Access HTC HTC Corporation HTML HyperText Markup Language HTTP Hypertext Transfer Protocol IB International Bureau iDEN Integrated Digital Enhanced Network ILEC Independent Local Exchange Carrier ITIF Information Technology & Innovation Foundation ITU International Telecommunication Union kbps Kilobits per Second LEC Local Exchange Carrier LLC Limited Liability Corporation LNP Local Number Portability LTE Long Term Evolution M&O Management and Operations M2M Machine-to-Machine MB Megabyte 163 Mbps Megabits per Second MEA Major Economic Area MHz Megahertz MIMO Multiple Input Multiple Output MMS Multimedia Messaging Service MOUs Minutes of use (average minutes of use per subscriber per month) MSA Metropolitan Statistical Area MSS Mobile Satellite Service MTA Major Trading Area MVNO Mobile Virtual Network Operator NCHS National Center for Health Statistics NFC Near-Field Communication NHIS National Health Interview Survey NIST National Institute of Standards and Technology NOI Notice of Inquiry NPA-NXX the first six digits of a ten-digit telephone number NPAC Number Portability Administration Center NPRM Notice of Proposed Rulemaking NRUF Numbering Report / Utilization Forecast NTCA National Telecommunications Cooperative Association NTIA National Telecommunications and Information Administration OBI Omnibus Broadband Initiative OET Office of Engineering & Technology OS Operating System PC Personal Computer PCS Personal Communications System PN Public Notice POPs population (people) PSTN Public Switched Telephone Network PUC Public Utility Commission R&D Research and Development R&O Report and Order RF Radio Frequency RIM Research in Motion RPM Revenue per Minute RSA Rural Service Area SEC Security and Exchange Commission SF 1 Summary File 1 SIM Subscriber Identity Module SMR Specialized Mobile Radio SMS Short Message Service TB Terabyte TDD Time Division Duplex TDM Time Division Multiplexing TDMA Time Division Multiple Access ULS Universal Licensing System UMTS Universal Mobile Telecommunications System US United States USB Universal Serial Bus USC United States Code USF Universal Service Fund VoIP Voice over Internet Protocol VZ Verizon WCDMA Wideband Code Division Multiple Access WCS Wireless Communications Service WiMAX Worldwide Interoperability for Microwave Access 164 WLAN Wireless Local Area Network WTB Wireless Telecommunications Bureau i Sue Marek, T-Mobile Gets ‘Jump’ on the Competition With New Handset Upgrade Program, FIERCEWIRELESS, July 10, 2013. ii Phil Goldstein, AT&T’s ‘Next’ Program Allows Smartphone, Tablet Upgrades Every 12 Months, FIERCEWIRELESS, July 16, 2013. iii Sue Marek, Verizon Gives Handset Upgrades Every 6 Months With New Edge Program, FIERCEWIRELESS, July 18, 2013. iv Phil Goldstein, Sprint Launches ‘One Up’ Handset Upgrade Program, Following Rivals, FIERCEWIRELESS, Sept.20, 2013; Phil Goldstein, Report: Sprint Set to Join Rivals With ‘One Up’ Handset Upgrade Program, FIERCEWIRELESS, Sept. 16, 2013. Sprint ended the ‘One Up’ program in January 2014 in conjunction with the launch of its Framily plans. Phil Goldstein, Sprint Kills ‘One Up’ Handset Upgrade Plan Less Than 4 Months After Launch, FIERCEWIRELESS, Jan. 13, 2014. v Tammy Parker, Sprint Kicks Off Framily, a Group-Based Calling Plan, FIERCEWIRELESS, Jan. 7, 2014; Phil Goldstein, AT&T, Verizon Tweak ‘Next’ and ‘Edge’ Handset Upgrade Programs, FIERCEWIRELESS, Jan. 21, 2014; Sprint, Learn More About Sprint Easy Pay, http://support.sprint.com/support/article/Learn_more_about_Sprint_Easy_Pay/487c6885-c186-4f0d-b325- 65b602af1de0?INTNAV=LP:RMB:01092014:EasyPay (visited Mar. 4, 2014). vi Craig Moffett et al., U.S. Wireless: The Rise of the Machines … and No, This Isn’t a Price War, MoffettNathanson Research, March 19, 2014, at 17; Sprint, Service Add-ons, http://shop.sprint.com/mysprint/shop/service_wall.jsp?_requestid=15971 (visited Apr. 10, 2014). vii Sue Marek, Verizon Gives Handset Upgrades Every 6 Months With New Edge Program, FIERCEWIRELESS, July 18, 2013. viii Phil Goldstein, AT&T, Verizon Tweak ‘Next’ and ‘Edge’ Handset Upgrade Programs, FIERCEWIRELESS, Jan. 21, 2014. ix Phil Goldstein, Confirmed: T-Mobile to Let ‘Jump’ Customers Upgrade More Often, Add Tablets to Program, FIERCEWIRELESS, Feb. 14, 2014. x Phil Goldstein, Verizon Tweaks Edge Handset Upgrade Program, as Analysts Predict Installment Plans to Boom, FIERCEWIRELESS, June 2, 2014. xi Phil Goldstein, Sprint Launches ‘One Up’ Handset Upgrade Program, Following Rivals, FIERCEWIRELESS, Sept.20, 2013; Phil Goldstein, Report: Sprint Set to Join Rivals With ‘One Up’ Handset Upgrade Program, FIERCEWIRELESS, Sept. 16, 2013. xii David W. Barden et al., The Rise of Installment Payment Plans and Implications for Wireless Carriers, Bank of America Merrill Lynch, Equity Research, Feb. 7, 2014, at 5; Phil Goldstein, , AT&T Targets T-Mobile With New ‘Mobile Share Value” No-Contract Pricing Options, FIERCEWIRELESS, Dec. 5, 2013. xiii David W. Barden et al., The Rise of Installment Payment Plans and Implications for Wireless Carriers, Bank of America Merrill Lynch, Equity Research, Feb. 7, 2014, at 5; Tammy Parker, Sprint Kicks Off Family, a Group- Based Calling Plan, FIERCEWIRELESS, Jan. 7, 2014; Sprint, Learn More About the Sprint Family Plan, http://support.sprint.com/support/article/title/315d1dfb-7fe2-4e5a-b633- 36db2003ecc8?INTNAV=LP:RMB:01092014:QA (visited Mar. 5, 2014). xiv Craig Moffett et al., U.S. Wireless: The Rise of the Machines … and No, This Isn’t a Price War, MoffettNathanson Research, March 19, 2014, at 17-18; Phil Goldstein, We’re Not Having a Wireless Price War Now -- But What Would One Look Like?, FIERCEWIRELESS, March 11, 2014. xv David W. Barden et al., The Rise of Installment Payment Plans and Implications for Wireless Carriers, Bank of America Merrill Lynch, Equity Research, Feb. 7, 2014, at 5; Phil Goldstein, Analyst: AT&T’s Cut in Mobile Share Family Pricing Targets T-Mobile, Verizon, FIERCEWIRELESS, Feb. 3, 2014; AT&T Launches Best-Ever Prices for Families on its Best-in-Class Network, Press Release, AT&T, Feb. 1, 2014. 165 xvi Debi Lewis, MORE Everything Gives Customers More From Their Wireless Plans, Press Release, Verizon Wireless, Feb. 13, 2014; Dan Seifert, Verizon’s New Rate Plans Offer More Data and Lower Prices, THEVERGE, Feb. 13, 2014; John C. Hodulik et al., Wireless Telecommunications: Higher EIP Take-Rates to Mask Pricing Pressure, UBS, Equity Research, March 10, 2014, at 5; Philip Cusick et al., Adding VZ to J.P.Morgan Analyst Focus List; Embrace the Flowback, J.P. Morgan, North America Equity Research, Feb. 27, 2014, at 11-14; Phil Goldstein, Analysts: Verizon’s ‘More Everything’ Plans Show No. 1 Carrier Isn’t Immune to Price War, FIERCEWIRELESS, Feb. 13, 2014. xvii Mike Dano, Verizon Cuts Prices to Match AT&T: 4 Lines With 10GB Now Costs $160, FIERCEWIRELESS, Apr. 2, 2014. xviii Phil Goldstein, C Spire Launches Promotion to Pay Off ETFs of Customers Who Switch Over, FIERCEWIRELESS, Apr. 1, 2014. xix Debi Lewis, More Savings Coming for Verizon Wireless Customers Beginning April 17, Press Release, Verizon Wireless, Apr. 14, 2014; Phil Goldstein, Verizon Cuts Pricing for ‘More Everything’ Subs Who Bring Their Own Phone, FIERCEWIRELESS, Apr. 14, 2014. xx C Spire, Better Than Ever – Shared Data With Overage Protection, http://www.cspire.com/promos/?q=shared- data (visited May 13, 2014). xxi Id.; C Spire Unveils New Customer-Inspired Shared Data Plans, Press Release, C Spire, May 5, 2014; Phil Goldstein, C-Spire Joins AT&T, Verizon in Offering 4 Lines, 10 GB of Data for $160, FIERCEWIRELESS, May 5, 2014. xxii Phil Goldstein, T-Mobile Kills Wireless Contracts, Will Launch iPhone 5 on April 12, FIERCEWIRELESS, Mar. 26, 2013; David W. Barden et al., The Rise of Installment Payment Plans and Implications for Wireless Carriers, Bank of America Merrill Lynch, Equity Research, Feb. 7, 2014, at 5. xxiii http://www.lightreading.com/t-mobile-kills-contracts-launches-lte-network/d/d-id/701971 xxiv Sprint Launches Unlimited Guarantee and New Unlimited, My Way Plan, Press Release, Sprint, July 11, 2013; Sue Marek, Sprint Guarantees Unlimited Data for Life, FIERCEWIRELESS, July 11, 2013; Michael Rollins, et al., Alert: Unlimited, Unlimited! New Sprint Brings New Rate Plans, Citigroup Global Markets Inc., Equity Research, July 11, 2013. xxv Sprint Launches Unlimited Guarantee and New Unlimited, My Way Plan, Press Release, Sprint, July 11, 2013; Sue Marek, Sprint Guarantees Unlimited Data for Life, FIERCEWIRELESS, July 11, 2013. xxvi Phil Goldstein, AT&T to Make All New Customers Sign Up for Mobile Share Plans, FIERCEWIRELESS, Oct. 11, 2013. xxvii Philip Cusick et al., AT&T Offers Off-Subsidy Contract Discounts, More Attractive NEXT Offerings and Larger Data Buckets, J.P. Morgan, Equity Research, Dec. 5, 2013, at 2; Simon Flannery et al., Quick Comment: New Mobile Share Value Plans Escalate Wireless Price Wars, Morgan Stanley, Equity Research, Dec. 6, 2013, at 4. xxviii Philip Cusick et al., AT&T Offers Off-Subsidy Contract Discounts, More Attractive NEXT Offerings and Larger Data Buckets, J.P. Morgan, Equity Research, Dec. 5, 2013, at 2 (noting that “a four-smartphone account using 10GB/month would pay 8% more …”); Craig Moffett et al., U.S. Wireless: The Rise of the Machines … and No, This Isn’t a Price War, MoffettNathanson Research, March 19, 2014, at 13-14 (noting that “for many families with low or moderate data use, the decreased cost of data will not be enough to offset the higher cost to add each smartphone line.”). xxix Phi Goldstein, Verizon Cuts Entry-Level Share Everything Plan Price by $20/Month, FIERCEWIRELESS, Jan. 21, 2014. xxx AT&T Launches Best-Ever Prices for Families on its Best-in-Class Network, Press Release, AT&T, Feb. 1, 2014; John C. Hodulik et al., Wireless Telecommunications: Higher EIP Take-Rates to Mask Pricing Pressure, UBS, Equity Research, Mar. 10, 2014, at 1-2; Dante D’Orazio, AT&T Tweaks Mobile Share Pricing to Significantly Discount 10GB or Larger Plans, THEVERGE, Feb. 1, 2014; Craig Moffett et al., U.S. Wireless: The Rise of the Machines … and No, This Isn’t a Price War, MoffettNathanson Research, Mar. 19, 2014, at 12,14. xxxi Mike Dano, Verizon Offering ‘Loyalty’ Plans to Retain Customers, Around $60/Month for Talk, Text and 2GH, FIERCEWIRELESS, Feb. 7, 2014; Craig Moffett et al., U.S. Wireless: The Rise of the Machines … and No, This Isn’t a Price War, MoffettNathanson Research, Mar. 19, 2014, at 13. 166 xxxii Philip Cusick et al., Adding VZ to J.P.Morgan Analyst Focus List; Embrace the Flowback, J.P. Morgan, North America Equity Research, Feb. 27, 2014, at 11-12; Phil Goldstein, Analysts: Verizon’s ‘More Everything’ Plans Show No. 1 Carrier Isn’t Immune to Price War, FIERCEWIRELESS, Feb. 13, 2014; Mike Dano, Verizon to Launch Cheaper ‘More Everything’ Plans Tomorrow, According to Reports, FIERCEWIRELESS, Feb. 12, 2014. xxxii Phil Goldstein, Analysts: Verizon’s ‘More Everything’ Plans Show No. 1 Carrier Isn’t Immune to Price War, FIERCEWIRELESS, Feb. 13, 2014. xxxiii T-Mobile Doubles Down on Flagship Simple Choice Plan With More 4G LTE Data, Tethering – And Unlimited International Texting, Press Release, T-Mobile, Mar. 10, 2014; Devindra Hardawar, T-Mobile Adds More Data to its Cheaper Plans – But Unlimited Data is Now $10 More, VENTURE BEAT, Mar. 7, 2014; Phil Goldstein, AT&T Cuts Mobile Share Value Prices, T-Mobile Adds Data to Simple Choice Plans, FIERCEWIRELESS, Mar. 10, 2014. xxxiii T-Mobile Doubles Down on Flagship Simple Choice Plan With More 4G LTE Data, Tethering – And Unlimited International Texting, Press Release, T-Mobile, Mar. 10, 2014; Thomas Gryta, T-Mobile Raises Price for Unlimited Plan, WALL STREET JOURNAL, Mar. 7, 2014; Devindra Hardawar, T-Mobile Adds More Data to its Cheaper Plans – But Unlimited Data is Now $10 More, VENTURE BEAT, Mar. 7, 2014; Phil Goldstein, T-Mobile CFO: $10 Price Increase on Unlimited Plans Needed to Recoup Investment, FIERCEWIRELESS, Mar. 10, 2014. xxxiv T-Mobile Doubles Down on Flagship Simple Choice Plan With More 4G LTE Data, Tethering – And Unlimited International Texting, Press Release, T-Mobile, Mar. 10, 2014; Thomas Gryta, T-Mobile Raises Price for Unlimited Plan, WALL STREET JOURNAL, Mar. 7, 2014; Devindra Hardawar, T-Mobile Adds More Data to its Cheaper Plans – But Unlimited Data is Now $10 More, VENTURE BEAT, Mar. 7, 2014; Phil Goldstein, T-Mobile CFO: $10 Price Increase on Unlimited Plans Needed to Recoup Investment, FIERCEWIRELESS, Mar. 10, 2014. xxxv AT&T Offers its Best Ever Pricing for Individual and Two Line Accounts on Nation’s Most Reliable 4G LTE Network, Press Release, AT&T, Mar. 10, 2014; David W. Barden et al., T/TMUS Prices Now Even, Prepay Battle Next, Bank of America Merrill Lynch, Equity Research, March 10, 2014, at 1; John C. Hodulik et al., Wireless Telecommunications: Higher EIP Take-Rates to Mask Pricing Pressure, UBS, Equity Research, March 10, 2014, at 1-2; Phil Goldstein, AT&T Cuts Mobile Share Value Prices, T-Mobile Adds Data to Simple Choice Plans, FIERCEWIRELESS, Mar. 10, 2014. xxxvi Phil Goldstein, Confirmed: U.S. Cellular Adds More Data to Shared Data Buckets, FIERCEWIRELESS, Mar. 13, 2014. xxxvii Phil Goldstein, T-Mobile Adds $40 ‘Simple Starter’ Plan, Promises More Moves Ahead, FIERCEWIRELESS, Apr. 9, 2014. xxxviii T-Mobile Abolishes Consumer Overages, Challenges Other Wireless Providers to Follow Suit, Press Release, T-Mobile, Apr. 14, 2014; Phil Goldstein, T-Mobile to Drop Domestic Overages for All Plans, Calls on AT&T, Verizon and Sprint to Follow Suit, FIERCEWIRELESS, Apr. 14, 2014. xxxix http://newsroom.sprint.com/news-releases/its-a-new-day-for-data-for-american-consumers.htm, 18 August, 2014 xl T-Mobile Introduces America’s Only Unlimited 4G LTE Family Plan, Press Release, December 9, 2014 http://newsroom.t-mobile.com/news/unlimited-family-plan.htm xli Sprint Launches Unlimited Guarantee and New Unlimited, My Way Plan, Press Release, Sprint, July 11, 2013; Sue Marek, Sprint Guarantees Unlimited Data for Life, FIERCEWIRELESS, July 11, 2013. xlii Phil Goldstein, AT&T Launches $5 Per Day Tablet Plan, Powers Tesla’s Wireless Services, FIERCEWIRELESS, Oct. 18, 2013. xliii Phil Goldstein, Updated: T-Mobile to Offer 200MB of Free Data for All Tablet Customers, FIERCEWIRELESS, Oct. 23, 2013; Phil Goldstein, T-Mobile CEO Sievert Reiterates 200 MB of Free Data for All Tablet Customers, FIERCEWIRELESS, Nov. 4, 2013. xliv Craig Moffett et al., U.S. Wireless: The Rise of the Machines … and No, This Isn’t a Price War, MoffettNathanson Research, March 19, 2014, at 10-11 (noting that qualifying tablet data plans start as low as $5 per month). xlv Phil Goldstein, AT&T to Offer Up to $450 in Credit to T-Mobile Customers Who Switch Over, FIERCEWIRELESS, Jan. 3, 2014; Phil Goldstein, AT&T Ends $450 Promoting Aimed at Wooing T-Mobile Customers, Cuts Aio Prices, FIERCEWIRELESS, Feb. 4, 2014; John C. Hodulik et al., Wireless 167 Telecommunications: Adding Fuel to the Fire with Further Pricing Changes, UBS, Equity Research, March 10, 2014, at 2. xlvi T-Mobile Delivers Contract Freedom for Families By Paying Off Early Termination Fees, Press Release, T- Mobile, Jan. 8, 2014; Sue Marek, T-Mobile Ends ETFs by Offering to Pay Up to $650 to Switchers, FIERCEWIRELESS, Jan. 8, 2014; John C. Hodulik et al., Wireless Telecommunications: Adding Fuel to the Fire with Further Pricing Changes, UBS, Equity Research, March 10, 2014, at 2; Craig Moffett et al., U.S. Wireless: The Rise of the Machines … and No, This Isn’t a Price War, MoffettNathanson Research, March 19, 2014, at 13. xlvii AT&T Ends $450 Promoting Aimed at Wooing T-Mobile Customers, Cuts Aio Prices, FIERCEWIRELESS, Feb. 4, 2014; John C. Hodulik et al., Wireless Telecommunications: Adding Fuel to the Fire with Further Pricing Changes, UBS, Equity Research, March 10, 2014, at 2. xlviii Sue Marek, Verizon Fuels Price War by Cutting $35 Activation Fee Until Feb. 17, FIERCEWIRELESS, Feb. 10, 2014. xlix Phil Goldstein, C Spire Launches Promotion to Pay Off ETFs of Customers Who Switch Over, FIERCEWIRELESS, Apr. 1, 2014. l Phil Goldstein, Verizon Gives 1 GB of Free Data to ‘More Everything’ Tablet Customers, FIERCEWIRELESS, Apr. 3, 2014. li For a Limited Time, Customers Who Switch a Number to Sprint on a FramilySM Plan Can Save up to $650, Press Release, Sprint, Apr. 4, 2014; Mike Dano, Confirmed: Sprint to Pay Up to $650 in ETFs for Framily Subscribers, Matches T-Mobile Offer Exactly, FIERCEWIRELESS, Apr. 4, 2014. lii lii U.S. Cellular Announces New Simple Connect No Contract Plans and a Retail Installment Option for Purchase of New Devices, Press Release, U.S. Cellular, Apr. 7, 2014; Phil Goldstein, U.S. Cellular Launches New No-Contract Plans, Installment Financing, FIERCEWIRELESS, Apr. 7, 2014. liii Day 2 of 3: T-Mobile Unleashes You to Experience Tablets Beyond the Wi-Fi Zone – For Free, Press Release, T-Mobile, Apr. 10, 2014; Phil Goldstein, T-Mobile Offers LTE-Enabled Tablets for Same Cost as Wi-Fi Tablets, FIERCEWIRELESS, Apr. 10, 2014; Phil Goldstein, T-Mobile Raises Prices on Some LTE Tablets After Promotion Ends, FIERCEWIRELESS, May 13, 2014. liv Day 2 of 3: T-Mobile Unleashes You to Experience Tablets Beyond the Wi-Fi Zone – For Free, Press Release, T-Mobile, Apr. 10, 2014; Phil Goldstein, T-Mobile Offers LTE-Enabled Tablets for Same Cost as Wi-Fi Tablets, FIERCEWIRELESS, Apr. 10, 2014. lv Flannery Morgan Stanley Research - T-Mobile US, Inc.(TMUS.N): Quick Comment: New Un-carrier Initiatives could Drive Further Disruption, June 19, 2014 lvi Phil Goldstein, AT&T Launches New Prepaid Service Under ‘Aio Wireless’ Brand, FIERCEWIRELESS, May 9, 2013. lvii Phil Goldstein, T-Mobile USA, MetroPCS Officially Tie Knot, FIERCEWIRELESS, May 1, 2013; Phil Goldstein, T-Mobile to Expand MetroPCS Footprint by 100M Pops, FIERCEWIRELESS, May 15, 2013; Phil Goldstein, T- Mobile Expands MetroPCS Footprint by 50M Pops, Adds $40 Rate Plan, FIERCEWIRELESS, July 25, 2013. lviii Phil Goldstein, Verizon Boosts Data Allotments in Prepaid Plans, FIERCEWIRELESS, May 16, 2013. lix Phil Goldstein, AT&T Adds LTE Service to Aio Wireless Prepaid Brand, FIERCEWIRELESS, June 20, 2013. lx Phil Goldstein, T-Mobile Expands MetroPCS Footprint by 50M Pops, Adds $40 Rate Plan, FIERCEWIRELESS, July 25, 2013 lxi Phil Goldstein, AT&T to Take Aio Wireless Prepaid Brand Nationwide in September, FIERCEWIRELESS, Aug. 30, 2013 lxii Phil Goldstein, AT&T’s Aio Reassures Customers as Leap Acquisition Approaches, FIERCEWIRELESS, Nov. 4, 2013. lxiii Phil Goldstein, Confirmed: T-Mobile, Target to Launch ‘Brightspot’ Prepaid Offerings, FIERCEWIRELESS, Oct. 3, 2013. lxiv Phil Goldstein, Sprint Launches Boost Mobile LTE Promotion, Cuts Price to $35/Month From $55/Month, FIERCEWIRELESS, Feb. 6, 2014. lxv Phil Goldstein, Verizon Cuts Prepaid Prices, Data Allotments With New ‘AllSet’ Plans, FIERCEWIRELESS, March 3, 2014. 168 lxvi Phil Goldstein, Sprint Replaces Sprint As You Go With Sprint Prepaid Brand: New plans at $45 and $60 for Smartphones, FIERCEWIRELESS, Mar. 14, 2014; John C. Hodulik et al., U.S. Wireless 411: Version 51, UBS, Equity Research, Mar. 18, 2014, at 7. lxvii Phil Goldstein, Sprint’s Boost, Virgin to Throttle Heavy Data Users to Slower Speeds Starting in May, FIERCEWIRELESS, March 20, 2014. lxviii Id.; Phil Goldstein, AT&T to Launch Renewed Cricket Prepaid Brand in Q2, FIERCEWIRELESS, Apr. 22, 2014. lxix Phil Goldstein, AT&T’s Aio Launches ‘Group Save’ Plans, Taking a Page From Sprint’s ‘Framily’, FIERCEWIRELESS, Apr. 17, 2014. lxx Phil Goldstein, Sprint’s Boost Cuts Prices, Introduces $40 Prepaid Plan in Challenge to T-Mobile, FIERCEWIRELESS, May 6, 2014. lxxi Phil Goldstein, AT&T to Launch Renewed Cricket Prepaid Brand at End of Q2, FIERCEWIRELESS, Apr. 22, 2014; Sue Marek, AT&T’s Revamped Cricket Will Take on T-Mobile’s MetroPCS With Aio-like Look, Rate Plans, FIERCEWIRELESS, May 18, 2014; Cricket’s Van Burskirk Talks Cricket’s Retail Presence, the Future of Muve, and More, FIERCEWIRELESS, June 3, 2014. lxxii Sue Marek, AT&T’s Revamped Cricket Will Take on T-Mobile’s MetroPCS With Aio-like Look, Rate Plans, FIERCEWIRELESS, May 18, 2014; Mike Dano, Rumor Mill: AT&T’s ‘New ‘Cricket’ Plans to Start at $40/Month, Speeds to Be Capped, FIERCEWIRELESS, May 8, 2014. lxxiii Mike Dano, AT&T Caps Cricket’s Peak Download Speeds to 8 Mbps on LTE, 4 Mbps on HSPA+, FIERCEWIRELESS, May 20, 2014. lxxiv Sue Marek, AT&T’s Revamped Cricket Will Take on T-Mobile’s MetroPCS With Aio-like Look, Rate Plans, FIERCEWIRELESS, May 18, 2014. lxxv Sue Marek, AT&T’s Revamped Cricket will Take on T-Mobile’s MetroPCS With Aio-like Look, Rate Plans, FIERCEWIRELESS, May 18, 2014; Mike Dano, Rumor Mill: AT&T’s ‘New Cricket’ Plans to Start at $40/Month, Speeds to Be Capped, FIERCEWIRELESS, May 8, 2014. lxxvi Sue Marek, AT&T’s Revamped Cricket Will Take on T-Mobile’s MetroPCS With Aio-like Look, Rate Plans, FIERCEWIRELESS, May 18, 2014. lxxvii Mike Dano, AT&T’s Cricket Offers Handset Financing on Prepaid Phones, Including Interest-Free Loans and Leases, FIERCEWIRELESS, May 22, 2014. lxxviii Phil Goldstein, T-Mobile cuts Prices, Adds Data to GoSmart Mobile Prepaid Plans, FIERCEWIRELESS, June 10, 2014. lxxix Id. lxxx Phil Goldstein, T-Mobile cuts Prices, Adds Data to GoSmart Mobile Prepaid Plans, FIERCEWIRELESS, June 10, 2014 lxxxi Kif Leswing. http://gigaom.com/2014/07/15/verizons-prepaid-plans-now-include-lte-data/ GIGAOM 14th July, 2014 lxxxii http://www.att.com/shop/wireless/plans/prepaidplans.html , Websites visited on 6/6/2014 lxxxiii https://www.cricketwireless.com/cell-phone-plans, Websites visited on 6/6/2014 lxxxiv http://www.sprint.com/landings/prepaid/; , http://www.boostmobile.com/shop/plans/monthly-unlimited- select/; Website visited on 6/6/2014, lxxxv lxxxv Website visited on 6/6/2014 lxxxvi http://prepaid-phones.t-mobile.com/prepaid-plans. Website visited on 6/6/2014 lxxxvii http://www.metropcs.com/metro/simpleplans. Website visited on 6/6/2014 lxxxviii lxxxviii https://www.gosmartmobile.com/ lxxxix http://prepaid-phones.t-mobile.com/prepaid-plans xc Website visited on 7/7/2014 xci As of 7/7/2014, http://www.cspire.com/shop_and_learn/plans/category_plan_landing.jsp?id=cat320003#prepaid. 169