Before the
Federal Communications Commission
Washington, D.C. 20554

In the Matter of
Inquiry Concerning the Deployment of
Advanced Telecommunications Capability to
All Americans in a Reasonable and Timely
Fashion, and Possible Steps to Accelerate Such
Deployment Pursuant to Section 706 of the
Telecommunications Act of 1996, as Amended
by the Broadband Data Improvement Act

)
)
)
)
)
)
)
)
)
)

GN Docket No. 15-191

COMMENTS OF AT&T

James P. Young
Christopher T. Shenk
Rishi P. Chhatwal
Marc A. Korman
Sidley Austin LLP
1501 K Street, N.W.
Washington, D.C. 20005
(202) 736-8000

Terri L. Hoskins
Gary L. Phillips
David L. Lawson
AT&T Services, Inc.
1120 20th Street, N.W.
Washington, D.C. 20036
(202) 457-2055

Attorneys for AT&T

September 15, 2015

 TABLE OF CONTENTS

INTRODUCTION AND SUMMARY ............................................................................................1
I.

THE PUBLICLY AVAILABLE FACTS EASILY CONFIRM THAT MOBILE
ADVANCED CAPABILITIES ARE BEING DEPLOYED IN A REASONABLE
AND TIMELY FASHION. .................................................................................................5

II.

THE COMMISSION’S PROPOSED ALTERNATIVE ANALYSIS IS
CONTRARY TO THE FACTS AND THE STATUTE. ...................................................12

III.

THE COMMISSION’S PROPOSED INQUIRY FOR MOBILE CAPABILITIES,
WHICH WOULD BE BASED ON METRICS SUCH AS ACTUAL SPEEDS
AND LATENCY, WOULD BE UNWORKABLE. ..........................................................14

CONCLUSION ..............................................................................................................................19

 Before the
Federal Communications Commission
Washington, D.C. 20554

In the Matter of
Inquiry Concerning the Deployment of
Advanced Telecommunications Capability to
All Americans in a Reasonable and Timely
Fashion, and Possible Steps to Accelerate Such
Deployment Pursuant to Section 706 of the
Telecommunications Act of 1996, as Amended
by the Broadband Data Improvement Act

)
)
)
)
)
)
)
)
)
)

GN Docket No. 15-191

COMMENTS OF AT&T
AT&T Inc. (“AT&T”), on behalf of itself and its affiliates, respectfully submits these
comments responding to the Commission’s Eleventh Broadband Progress Notice of Inquiry.1
INTRODUCTION AND SUMMARY
Section 706 requires the Commission to determine each year “whether advanced
telecommunications capability is being deployed to all Americans in a reasonable and timely
fashion.”2 In its most recent Broadband Progress Report, the Commission instead defined
“advanced telecommunications capability” based on the extreme bandwidth levels that
households with multiple persons simultaneously engaged in bandwidth-intensive entertainment

1

Eleventh Broadband Progress Notice of Inquiry, Inquiry Concerning the Deployment of
Advanced Telecommunications Capability to All Americans in a Reasonable and Timely
Fashion, and Possible Steps to Accelerate Such Deployment Pursuant to Section 706 of the
Telecommunications Act of 1996, as Amended by the Broadband Data Improvement Act, GN
Docket No. 15-191, FCC 15-101 (rel. Aug. 7, 2015) (“Notice”).

2

47 U.S.C. § 1302(b).

 and other activities might want,3 rather than focusing, as the statute requires, on the minimum
level of service consumers would need to access the most common and “regularly use[d]”
services.4 Applying this erroneous standard, the Commission defined advanced capabilities as
requiring at least 25 Mbps download speeds. Then, based on a finding that some 17 percent of
Americans live in areas where that capability is not currently available, and ignoring that this
percentage was declining rapidly, the Commission concluded that advanced telecommunications
capability is not occurring in the U.S. broadly or quickly enough.
The Commission now proposes to extend this unlawful approach to the mobile context
and, in the process, unreasonably narrow the definition of advanced telecommunications
capability even further. Specifically, the Commission asks whether a consumer lacks access to
advanced telecommunications capability unless she has access to both a 25 Mbps fixed
broadband service and a 10 Mbps mobile broadband service.5
AT&T does not object to consideration of mobile broadband services in an inquiry as to
whether advanced telecommunications capability is being reasonably and timely deployed in the
United States. But any such inquiry would have to begin and end with the obvious fact that the
wireless industry has deployed the most advanced technology available – LTE – on a nationwide
basis. Indeed, the United States is the world leader in deployment of mobile broadband services,
3

2015 Broadband Progress Report and Notice of Inquiry on Immediate Action to Accelerate
Deployment, Inquiry Concerning the Deployment of Advanced Telecommunications Capability
to All Americans in a Reasonable and Timely Fashion, and Possible Steps to Accelerate Such
Deployment Pursuant to Section 706 of the Telecommunications Act of 1996, as Amended by the
Broadband Data Improvement Act, GN Docket No. 14-126, ¶ 9 (Feb. 4, 2015) (“2015
Broadband Progress Report”).
4

Verizon v. FCC, 740 F.3d 623, 641 (D.C. Cir. 2014) (noting that Commission previously
increased threshold to 4 Mbps to establish a “threshold more appropriate to current consumer
behavior and expectations” and that was “enough” to permit what “consumers now regularly
use”).
5

See Notice ¶ 30.
2

 and carriers are investing tens of billions of dollars every year to increase service quality, speed,
capacity, and coverage. As the Commission’s own annual reports to Congress on the state of the
mobile wireless services industry acknowledge, more than 98 percent of Americans live in areas
where at least two carriers offer 4G LTE service and more than 80 percent have access to four.6
Given that the industry has achieved ubiquitous deployment of the most “advanced” capabilities,
the only legitimate conclusion is that mobile advanced services are being reasonably and timely
deployed.
Instead of acknowledging this obvious conclusion, the Commission proposes a granular
analysis of mobile capabilities based on speed, latency, and other metrics that are arbitrary and
unworkable. For example, the 10 Mbps speed minimum the Commission proposes for mobile
broadband is literally plucked from thin air; the Commission offers no explanation whatsoever
for how it arrived at that proposed benchmark, much less how it can be reconciled with the
Commission’s (incorrect) hypothesis that consumers use mobile services primarily for lowbandwidth uses like social media and turn-by-turn directions, rather than video. But even if the
Commission acknowledges that data usage for video has exploded on mobile networks in recent
years, it could not justify a 10 Mbps speed benchmark. For example, both YouTube and Netflix
report that their high quality streaming can be done with download speeds far less than 10 Mbps
(1 Mbps for YouTube and 5 Mbps for Netflix).7 And even these metrics ignore that many video
providers offer smaller file sizes optimized for the much smaller screens used with mobile
6

Seventeenth Report, 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, 29 FCC Rcd. 15311, ¶ 59 (Table
III.A.2) (2014) (“Seventeenth Mobile Competition Report”).
7

See
Netflix,
“Internet
Connection
Speed
Recommendations,”
https://help.netflix.com/en/node/306; see also YouTube, YouTube Help, “System
Requirements,” https://support.google.com/youtube/answer/78358?hl=en.
3

 devices. Moreover, unlike wireline services which may be shared by multiple users, mobile
services are typically used by one person at a time. In short, there is no real world justification
for a finding that advanced services over mobile require at least 10 Mbps download speeds.
To make matters worse, the Notice proposes to develop some method of mapping
“deployment” of advanced mobile capabilities that turns on geographically granular data
concerning download and upload speed.

But as explained below, the Commission has

previously “excluded mobile” from its Section 706 analysis precisely “due to concerns about the
quality and reliability of the data for these services,”8 and that lack of data still exists. The
proposal to fill this data gap with Form 477 data will not work. Form 477 contains only
“minimum advertised speeds” for mobile services and the Commission itself has emphasized that
it is “actual speeds” that matter. The use of minimum advertized speeds from Form 477 could
only systematically undercount locations where actual download and upload speeds exceed the
threshold for advanced capabilities. Nor could the Commission fill the gap with the alternative
data sources suggested in the Notice, which existed when the Commission previously found a

8

Notice ¶ 5. See also Seventh Broadband Progress Report and Order on Reconsideration,
Inquiry Concerning the Deployment of Advanced Telecommunications Capability to All
Americans in a Reasonable and Timely Fashion, and Possible Steps to Accelerate Such
Deployment Pursuant to Section 706 of the Telecommunications Act of 1996, as Amended by the
Broadband Data Improvement Act, 26 FCC Rcd. 8008 ¶ 26 (2011) (“Seventh Broadband
Progress Report”) (“We do not draw conclusions…about mobile wireless services due to our
concern that these data do not accurately reflect where mobile wireless subscribers actually are
able to obtain services that meets the broadband performance threshold.”); see also Eighth
Broadband Progress Report, Inquiry Concerning the Deployment of Advanced
Telecommunications Capability to All Americans in a Reasonable and Timely Fashion, and
Possible Steps to Accelerate Such Deployment Pursuant to Section 706 of the
Telecommunications Act of 1996, as Amended by the Broadband Data Improvement Act, 27 FCC
Rcd. 10342 ¶ 35 (2012) (“Eighth Broadband Progress Report”) (“[W]e do not include the
mobile data in our statutory finding in this report for two reasons. First, as detailed below, we
have concerns that the available data sources for measuring mobile broadband may overstate
deployment to a significant degree.”).
4

 lack of adequate data sources and which do not provide the kind of granular or comprehensive
data the Commission would need for its proposed analysis.
The Commission’s proposal to develop metrics for various additional under-the-hood
metrics, such as latency and consistency, would be even more unworkable and arbitrary. These
metrics have trade-offs (making one metric better can make others worse), and providers
carefully balance those factors to provide high quality user experiences. Thus, the focus must be
on the overall user experience, not individual performance metrics like latency.
In the remainder of these comments, we demonstrate that: (1) the publicly available data
easily confirm that the most advanced mobile capabilities available are being deployed in a
reasonable and timely fashion; (2) there is no basis for the proposal that consumers require
mobile broadband speeds of at least 10 Mbps to have access to advanced telecommunications
capability; and (3) measuring mobile broadband “deployment” by means of geographically
granular metrics such as speed, latency, and “consistency,” as the Commission proposes, is
unworkable for mobile technologies, because the Commission does not have the data sources
needed to conduct any such analysis.
I.

THE PUBLICLY AVAILABLE FACTS EASILY CONFIRM THAT MOBILE
ADVANCED CAPABILITIES ARE BEING DEPLOYED IN A REASONABLE
AND TIMELY FASHION.
Section 706 requires the Commission to answer a simple question: “whether advanced

telecommunications capability is being deployed to all Americans in a reasonable and timely
fashion.”9

“Advanced telecommunications capability” is defined as “high-speed, switched,

broadband telecommunications capability that enables users to originate and receive high-quality

9

47 U.S.C. § 1302(b).
5

 voice, data, graphics, and video telecommunications using any technology.”10 AT&T does not
object to consideration of mobile broadband services in a Section 706 inquiry, but if the
Commission is going to add mobile capabilities and technologies to the analysis, the
Commission should answer the question Congress asked. And for mobile capabilities, the
answer is easy: publicly available data amply confirm that U.S. mobile deployment far exceeds
Section 706’s “reasonable and timely” standard.
In the context of mobile capabilities, the most “advanced” technology that any nation
could reasonably deploy today is LTE.

The U.S. industry is obviously deploying those

capabilities in a “reasonable and timely” manner, because virtually “all Americans” now have
access to multiple LTE providers. As the Commission and the White House have rightly touted,
the U.S. leads the world in mobile broadband deployment and usage, and by a wide margin.
That is why a senior official in Chairman Wheeler’s office recently characterized U.S. “mobile
broadband networks” as “the envy of the world.”11 The experts at 4G Americas agree that “[i]n
terms of market share, penetration and coverage, the U.S. leads the world and remains at the
head of the class in technology and spectrum innovation.”12 And U.S. mobile deployment is
clearly “timely”: a recent White House Report noted that U.S. providers met the President’s

10

Id. § 1302(d)(1).

11

Remarks of Gigi B. Sohn, Counselor to FCC Chairman Wheeler, Moving Towards a Gigabit
State (May 4, 2015), available at https://www.fcc.gov/document/remarks-gigi-b-sohn-movingtowards-gigabit-state.
12

Year-End 2014: Nearly Half a Billion LTE Connections WorldWide, 4G Americas reports
substantial gains for LTE in North America, 4G Americas (March 11, 2015),
http://www.4gamericas.org/en/newsroom/press-releases/year-end-2014-nearly-half-billion-lteconnections-worldwide.
6

 challenge to deploy 4G technology to 98 percent of Americans “nearly two years ahead of
schedule.”13
As that White House assessment confirms, the U.S. wireless industry is deploying
advanced mobile capabilities at breakneck speed. At the turn of this decade, LTE was still on the
standard-setting bodies’ drawing boards. But as the Commission recently found in its Mobile
Competition Report, by January 2014, 98.5 percent of Americans were covered by LTE.14
AT&T’s facilities-based LTE network now covers about 96 percent of the U.S. population,15 and
AT&T covers most other Americans (totaling more than 99 percent) through roaming
agreements with other LTE providers. Although AT&T’s efforts alone would satisfy Section
706’s standard for deployment to all Americans, other carriers likewise have deployed extensive
LTE networks.

Verizon reports that its LTE network “cover[s] more than 98 percent of

Americans.”16 T-Mobile recently stated that its facilities-based LTE network covers about 290
million Americans (about 90 percent), and it expects to cover 300 million Americans by the end
of 2015.17 Sprint has deployed an LTE network that covers about 280 million POPs (about 87

13

White House, Office of the Press Secretary, Fact Sheet: Next Steps in Delivering Fast
Affordable Broadband (March 23, 2015), available at https://www.whitehouse.gov/the-pressoffice/2015/03/23/fact-sheet-next-steps-delivering-fast-affordable-broadband.
14

Seventeenth Mobile Competition Report ¶ 59 (Table III.A.2).

15

See AT&T, “Network,” http://www.att.com/network/en/index.html (“Overall, AT&T covers
over 320 million people [out of about 321 million people in the U.S.] with its voice and data
services, with over 308 million people covered by our 4G LTE network”).

16

Verizon,
“Why
It’s
Good
to
Be
a
Verizon
Customer,”
http://www.verizonwireless.com/mobile-living/network-and-plans/why-verizon-customerexclusives/.
17

T-Mobile, T-Mobile Investor Factbook, Q2 2015, at 2, available at http://investor.tmobile.com/Cache/1500074332.PDF?Y=&O=PDF&D=&fid=1500074332&T=&iid=4091145.
7

 percent).18 Regional and local providers have likewise been blanketing their service areas with
LTE. For example, U.S. Cellular reports that “94 percent of [its] customers have access to 4G
LTE speeds”19 and that it “plans to cover essentially its entire customer base with LTE by the
end of [2015],”20 and nTelos is investing $175 million to roll out LTE across its footprint.21
Thus, as the Commission’s most recent Mobile Competition Report confirms, virtually every
American has access to multiple LTE broadband providers.22
The U.S.’s large lead in LTE deployment has translated into a similarly large lead in LTE
adoption and usage. The U.S. has the most LTE subscribers (158 million connections) in the
world, despite having the third lowest amount of spectrum dedicated to LTE.23 As reported by
CTIA, by the end of 2014, “LTE penetration in the U.S. [was] at nearly 50%,” compared to only
13 percent of mobile connections in Western Europe and 10 percent in the Asia Pacific region.24
About 72 percent of all smartphones in North America (and on AT&T’s network) are LTE-

18

Sprint, Presentation, FY4Q14 Results Conference Call, at 9, May 5, 2015,
http://investors.sprint.com/Cache/1500071437.PDF?Y=&O=PDF&D=&FID=1500071437&T=
&IID=4057219 (“Sprint Conference Call”).
19

See U.S. Cellular Website, http://www.uscellular.com/about/index.html.

20

Phil Goldstein, U.S. Cellular to expand LTE network cover 98% of its customers by end of
2015, Fierce Wireless, April 1, 2015, http://www.fiercewireless.com/story/us-cellular-expandlte-network-cover-98-its-customers-end-2015/2015-04-01.
21

nTelos, Press Release, nTelos Announces “4G for All” Network Expansion Plan, First LTE
Launch of 2015, Fierce Wireless, Feb. 2, 2015, http://www.fiercewireless.com/pressreleases/ntelos-announces-4g-all-network-expansion-plan-first-lte-launch-2015.
22

Seventeenth Mobile Competition Report ¶ 51 (Table III.A.2).

23

See Thomas K. Sawanobori and Dr. Robert Roche, CTIA – The Wireless Association, Mobile
Data Demand: Growth Forecasts Met, June 22, 2015, at 4, available at
http://www.ctia.org/docs/default-source/default-document-library/062115mobile-data-demandswhite-paper.pdf (“Mobile Data Demand: Growth Forecasts Met”).
24

See Comments of CTIA – The Wireless Association, Wireless Telecommunications Bureau
Seeks Comment on the State of Mobile Wireless Competition, WT Docket No. 15-125, at 66
(June 29, 2015) (“Comments of CTIA”).
8

 compatible, compared to only 51 percent for Western Europe.25 More than 72 percent of data
traffic in the U.S. is LTE traffic,26 compared to only 31 percent for Western Europe.27 And, U.S.
mobile end users generated an average of 1,503 MB of mobile data traffic per month, compared
to only 373 MB for Central and Eastern Europe, and 273 MB for the Asia-Pacific region.28
Indeed, as a result of the prevalence and quality of mobile broadband in the U.S.,
Americans are increasingly using advanced mobile broadband capabilities to access the same
applications as wireline services, especially for the “voice, data, graphics, and video” services
that define “advanced services” under Section 706. A recent analysis by Cisco reports that
“viewing of mobile video offerings . . . accounted for 60 percent of all mobile data traffic at the
end of 2014,”29 and that video will be 75 percent of the United States’ mobile data traffic by
2019.30 “YouTube now reaches more of the 18-49 year-old demographic – on just mobile
devices – than any cable television network,” and mobile device viewing averaged about 2.8

25

See id.; see also AT&T Investor Briefing No. 289, 2nd Quarter 2015, July 23, 2015, available
at http://www.att.com/Investor/Earnings/2q15/ib_final_2q15.pdf.
26

Coleman Bazelon and Guilia McHenry, Substantial Licensed Spectrum Deficit (2015-2019):
Updating the FCC’s Mobile Data Demand Projections, at 4-5 (June 23, 2015), available at
http://www.ctia.org/docs/default-source/default-document-library/bazelon_mchenry_spectrumdeficit_2015-06-23.pdf.
27

Comments of CTIA, at 66.

28

Id. at 68.

29

Mobile Data Demand: Growth Forecasts Met, at 5 (citing Cisco, VNI Mobile Forecast
Highlights, 2014-2019 (United States – Mobile Applications),
http://www.cisco.com/assets/sol/sp/vni/forecast_highlights_mobile/index.html#~Country).

30

See Cisco, VNI Mobile Forecast Highlights, 2014-2019 (United States – Mobile Applications),
available
at
http://www.cisco.com/c/dam/assets/sol/sp/vni/forecast_highlights_mobile/index.html#~Country;
see also Mobile Data Demand: Growth Forecasts Met, at 5.
9

 hours a day in 2015 (compared to only 2.4 hours per day for desktop computers).31 Half of
AT&T’s “mobile network traffic is video now, and [it is] really growing fast.”32 And Verizon
just launched a trial of a new over-the-top mobile service video service called “Go90” “primed
with around 100 to 200 hours of exclusive content from online video networks” and “popular
shows from Comedy Central, Food Network, ESPN, NFL Network, and Discovery.”33 As to
data and graphics, the Commission itself points out that “[m]obile broadband has become
increasingly important for accessing websites, navigating during travel, connecting on social
media, communicating with family and friends, receiving timely news updates, and obtaining
entertainment.”34
These trends in mobile advanced capabilities are expected to continue. A recent report
shows that Americans now spend more of their Internet time on mobile devices (about 60
percent) than on wireline devices (about 40 percent).35 Moreover, an increasing portion of
Americans are cutting their wireline usage altogether, relying solely on mobile devices for
broadband and voice services. As one recent report shows, “21 percent [of millennials] are no

31

Mary Meeker, Internet Trends 2015 – Code Conference, KPCB, at 24 (May 27, 2015),
available at http://www.kpcb.com/internet-trends; see also Mobile Data Demand: Growth
Forecasts Met, at 5.
32

Jonathan Vanian, AT&T CEO: “Half of our mobile network traffic is video,” Fortune (June
12, 2015), available at http://fortune.com/2015/06/12/att-randall-stephenson-video-traffic.
33

Phil Goldstein, Verizon launches Go90 mobile OTT video service as trial for some VZW subs;
full
launch
coming
soon,
FierceWireless,
Sept.
8,
2015,
http://www.fiercewireless.com/story/verizon-launches-go90-mobile-ott-video-service-trial-somevzw-subs-full-lau/2015-09-08.
34

Notice ¶ 8.

35

Sarah Perez, Majority Of Digital Media Consumption Now Takes Place In Mobile Apps,
TechCrunch, Aug. 21, 2014, http://techcrunch.com/2014/08/21/majority-of-digital-mediaconsumption-now-takes-place-in-mobile-apps/ (“mobile usage as a whole accounts for 60% of
time spent, while desktop-based digital media consumption makes up the remaining 40%.”); see
also Mobile Data Demand: Growth Forecasts Met, at 5.
10

 longer using desktop computers to go online.”36 Similar trends exist for the 55-years-and-older
consumer segment.37
In addition, providers of mobile broadband in the U.S. have spent billions of dollars to
develop and deploy the next generation of mobile services. For example, AT&T and other
providers have begun deployment of next-generation LTE-Advanced mobile broadband
technologies,38 which enable even faster and more reliable connections through cutting-edge
transmission technologies like carrier aggregation39 and “multi-layer MIMO.”40

Similarly,

AT&T and others are deploying the next generation of mobile broadband voice services,
including Voice over LTE (“VoLTE”).41 AT&T, Verizon Wireless, and T-Mobile have all
deployed VoLTE, and each carrier has used VoLTE to provide superior voice calls via high
definition voice services (HD Voice).42 At the same time, providers’ investments in the rapid

36

Kate Dreyer, Mobile Internet Usage Skyrockets in Past 4 Years to Overtake Desktop as Most
Used
Digital
Platform,
comScore,
April
13,
2015,
https://www.comscore.com/Insights/Blog/Mobile-Internet-Usage-Skyrockets-in-Past-4-Years-toOvertake-Desktop-as-Most-Used-Digital-Platform.
37

Id.

38

4G Americas, “LTE-Advanced,”
education/lte-advanced/.

http://www.4gamericas.org/en/resources/technology-

39

See Dr. John Saw, Sprint, Chief Network Officer, “The Sprint Network is Getting Better Every
Day,” May 20, 2015, http://newsroom.sprint.com/blogs/sprint-perspectives/the-sprint-networkis-getting-better-every-day.htm; Matt Hamblen, Sprint to rollout LTE Advanced to Chicago
area, Computer World, March 30, 2015, http://www.computerworld.com/article/2903874/sprintto-rollout-lte-advanced-to-chicago-area.html (Sprint begins its LTE advanced rollout with
Chicago in March 2015).

40

Phil Goldstein, Sprint to bring LTE carrier aggregation, multi-layer MIMO to Chicago in
investment push, Fierce Wireless, March 31, 2015, http://www.fiercewireless.com/story/sprintbring-lte-carrier-aggregation-multi-layer-mimo-chicago-investment-pu/2015-03-31.
41

Sue Marek, The voice renaissance: VoLTE, HD Voice and Wi-Fi calling bring innovation to
voice, Fierce Wireless, Feb. 20, 2015, http://www.fiercewireless.com/story/voice-renaissancevolte-hd-voice-and-wi-fi-calling-bring-innovation-voice/2015-02-20.
42

Id.
11

 deployment of these advanced mobile technologies have enabled dramatic growth throughout the
wireless ecosystem, as evidenced by the enormous growth in the “Internet of Things.”43
In short, the answer to the question whether advanced mobile capabilities are being
deployed in the U.S. on a reasonable and timely basis is obviously yes. That is all the more
evident given that Section 706 does not merely require a report to Congress; if the Commission
concludes that deployment has not been reasonable or timely, Section 706 requires it to “take
immediate action to accelerate deployment of such capability.”44 It would be absurd on these
facts to conclude that deployment has fallen below the Section 706 “reasonable and timely”
standard or that special Commission intervention is somehow necessary to “accelerate” the
industry’s efforts. In the end, the only possible conclusion is the one dictated by common sense:
if the United States has achieved ubiquitous deployment of the most advanced technology
available and leads the rest of the world by a wide margin, how can its deployment not be
“reasonable” or “timely”?
II.

THE COMMISSION’S PROPOSED ALTERNATIVE ANALYSIS IS CONTRARY
TO THE FACTS AND THE STATUTE.
Although today’s nationwide deployment of LTE easily satisfies the statutory standard of

“reasonable” and “timely” deployment, the Commission proposes an artificial and abstract
inquiry that would lead to the facially implausible conclusion that advanced mobile capabilities
are not being deployed in a reasonable or timely manner. Indeed, the Commission’s proposed 10
Mbps/1 Mbps threshold for mobile capabilities is out of left field. The Notice provides no
analysis to support this figure at all; it literally does nothing more than ask, “how about 10
43

Deloitte, “United States expands global lead in mobile broadband: How policy actions could
enhance or imperil America’s mobile broadband competitiveness,” at 17 (Sept. 2014),
http://www2.deloitte.com/content/dam/Deloitte/us/Documents/technology-mediatelecommunications/us-tmt-mobile-index-09262014.pdf.
44

47 U.S.C. § 1302(b).
12

 Mbps?”45 Under the Commission’s theory, however, consumers use mobile broadband for
relatively non-bandwidth-intensive tasks like accessing Facebook or using turn-by-turn
directions. The Commission makes no attempt to explain why 10 Mbps download speeds would
be necessary to use such services. As demonstrated above, more than half of all mobile traffic
today is related to video, thus vividly demonstrating that today’s mobile broadband services are
more than capable of providing even the most bandwidth intensive services, making any speed
benchmark unnecessary for assessing mobile services under any Section 706 analysis.
Moreover, 10 Mbps would be far too high a benchmark for mobile broadband. Both YouTube
and Netflix report that their high definition streaming can be done with download speeds far less
than 10 Mbps (1 Mbps and 5 Mbps, respectively).46 Even these metrics ignore that many video
services offer smaller file sizes optimized for the much smaller screens used with mobile
devices.47 And, unlike wireline services which may be shared by multiple users, mobile services
are typically used by one person at a time.
More fundamentally, however, the Commission is repeating the mistake it made last year
in redefining the fixed wireline download speed thresholds. Congress intended Section 706 to
act as a kind of emergency alarm system; the Commission is supposed to determine whether
deployment is lagging materially behind what the industry could reasonably achieve, and if so,
the statute requires Commission to try to accelerate deployment by removing barriers to
45

Notice ¶ 30 (“would a mobile speed benchmark of 10 Mbps/1 Mbps be appropriate?”).

46

See Netflix, “Internet Connection Speed Recommendations,”
https://help.netflix.com/en/node/306; see also YouTube, YouTube Help, “System
Requirements,” https://support.google.com/youtube/answer/78358?hl=en.
47

See, e.g., Jan Ozer, The State of Video Codecs 2015, StreamingMedia.com (March 2015),
http://www.streamingmedia.com/Articles/Editorial/Featured-Articles/The-State-of-VideoCodecs-2015-102806.aspx (“In 2014, some companies began offering technologies that claim to
reduce the delivery data rate of H.264 by as much as 50 percent with minimal impact on visual
quality.”).
13

 investment and promoting competition. The Commission, however, has twisted the statute so
that advanced telecommunications capability is defined to embody the Commission’s “stretch”
goals, effectively judging the industry in relation to download speeds that the most data-intensive
users would desire. That reading of the statute is unreasonable, because it essentially guarantees
that the Commission will always find deployment falling short of what is reasonable and timely,
thus enabling the Commission to perpetually declare for itself authorities that it might not
otherwise have. The Commission and the D.C. Circuit have treated the statutory definition as
including only the minimum bandwidth necessary to enable “the most popular forms” of Internet
uses and the services that consumers “regularly” use,48 and the Commission should return to that
standard.
III.

THE COMMISSION’S PROPOSED INQUIRY FOR MOBILE CAPABILITIES,
WHICH WOULD BE BASED ON METRICS SUCH AS ACTUAL SPEEDS AND
LATENCY, WOULD BE UNWORKABLE.
Even if it made sense to adopt the type of alternative analysis the Commission proposes

in the Notice, any such analysis would be completely unworkable. The Commission essentially
proposes to develop some method of mapping “deployment” that turns on geographically
granular data concerning speed and various other under-the-hood metrics (e.g., latency). The
Commission simply does not have the data sources needed to perform this type of analysis. The
performance of wireless networks varies substantially from location to location and from time to
48

Report, Inquiry Concerning the Deployment of Advanced Telecommunications Capability to
All Americans in a Reasonable and Timely Fashion, and Possible Steps to Accelerate Such
Deployment Pursuant to Section 706 of the Telecommunications Act of 1996, as Amended by the
Broadband Data Improvement Act, 14 FCC Rcd. 2398 ¶ 20 (1999) (choosing original 200 Kbps
threshold because it was “enough to provide the most popular forms of broadband—to change
web pages as fast as one can flip through the pages of a book and to transmit full-motion video”)
(emphasis added); Verizon v. FCC, 740 F.3d 623, 641 (D.C. Cir. 2014) (noting that Commission
increased threshold to 4 Mbps to establish a “threshold more appropriate to current consumer
behavior and expectations” and that was “enough” to permit what “consumers now regularly
use”).
14

 time, depending on a variety of factors (e.g., available spectrum, propagation, terrain, buildings
materials, concentration of users, peak usage times, end-user device capabilities, services being
used by end-users, and so on), and thus there would be enormous challenges to developing and
interpreting the sorts of speed and other metrics as proposed in the Notice. In prior reports, the
Commission has “excluded mobile . . . broadband service due to concerns about the quality and
reliability of the data for these services,”49 and those problems still exist.
First, the Commission does not have the data to measure “deployment” as a granular map
showing where actual speeds exceed 10 Mbps. The “minimum advertised speeds” reported by
providers on Form 477 certainly would not be of any assistance.50 As the Commission itself
acknowledges, what matters is “actual speeds,”51 but the advertised minimum available speeds
will always be lower than average speeds that consumers experience, thus providing misleading
data that will systematically undercount available download and upload speeds for mobile
services. For example, AT&T advertises a minimum download speed of 5 Mbps for 4G LTE,
which is reflected on AT&T’s Form 477.52 But according to RootMetrics, AT&T’s median
download speed was between 10-20 Mbps in 79 metro markets and 20 Mbps or faster in 11
markets in the first half of 2015, far higher than the 5 Mbps minimum advertised download

49

Notice ¶ 5. See also Seventh Broadband Progress Report ¶ 26 (“We do not draw
conclusions…about mobile wireless services due to our concern that these data do not accurately
reflect where mobile wireless subscribers actually are able to obtain services that meets the
broadband performance threshold.”); see also Eighth Broadband Progress Report ¶ 35 (“[W]e
do not include the mobile data in our statutory finding in this report for two reasons. First, as
detailed below, we have concerns that the available data sources for measuring mobile
broadband may overstate deployment to a significant degree.”).

50

Notice ¶ 5.

51

Id. ¶ 30.

52

See AT&T, Broadband Information, http://www.att.com/gen/public-affairs?pid=20879.
15

 speed (and far higher than the Commission’s proposed benchmark of 10 Mbps).53 Consequently,
if the Commission were to use AT&T’s minimum advertised speed from Form 477, the
Commission would incorrectly conclude that AT&T’s speeds are below the 10 Mbps threshold
(or whatever arbitrary level the Commission adopts) needed to qualify as an advanced capability.
Moreover, measuring minimum speeds would be inconsistent with how the Commission assesses
fixed broadband, which has been based on averages.54
The Commission acknowledges that Form 477 data is inadequate and asks whether “other
data sources” can be used to fill the gaps.55 The Notice identifies various candidates, including
Measuring Mobile Broadband America, Ookla, RootMetrics, Google M-Lab, and CalSPEED.56
To be sure, each of these data sources has independent value for consumers by allowing them to
obtain a measure of speed (and sometimes latency) at a specific location and time of day. But
these data sources do not provide the sort of comprehensive information that could be relied
upon by the Commission to draw conclusions about mobile broadband speeds at granular
geographic areas throughout the U.S.
Most of these data sources rely on customer-performed testing using the mobile
applications provided by these entities. As a result these data do not account for important
factors that may affect the speeds and other metrics. For example, they do not account for
whether the user’s device was running other applications using the broadband connection at the
same time the speed test was being conducted, whether the device supported the fastest available

53

See Dave Anderson, Mobile Network Performance in the US: First Half 2015, RootMetrics,
Aug. 18, 2015, http://www.rootmetrics.com/us/blog/special-reports/2015-1h-national-us
(“RootMetrics First Half 2015 Performance”).
54

2015 Broadband Progress Report ¶ 44.

55

Notice ¶ 5.

56

Id. ¶¶ 7, 67.
16

 speeds (e.g., whether it supported the relevant LTE bands and carrier aggregation), whether the
operating system supported the most advanced available technologies, whether the testing
occurred in a unique non-representative environment (e.g., tests conducted underground in a
parking garage would typically produce slower speeds than at street level). Moreover, many of
these data sources are not available at the geographic granularity needed for the proposed Section
706 analysis.

For example, CalSPEED only measures broadband speed in one state and

RootMetrics only provides statistics for a subset of geographic areas (nationwide, statewide, the
125 largest metro areas, 50 busiest airports, special event venues, busy transit stations, and
certain university campuses).57
In addition to the lack of data, the Commission’s proposal to assess various additional
under-the-hood metrics, such as latency and consistency, would be even more unworkable. In
any given wireless network, characteristics such as “latency” reflect a careful balance in deciding
how to provide customers the desired mobile broadband experience. Such metrics therefore
cannot be legitimately examined in isolation. The proposal to examine latency illustrates the
problem. Latency interacts with multiple other network performance metrics, including packet
loss. One way to decrease latency is to reduce the size of the buffers in network routers, thus
reducing the wait time for each packet when it arrives at the router. But reducing buffer sizes
results in increased packet loss because packet loss typically occurs when a router’s buffer is
full. Excessive packet loss can substantially reduce speed (because packets must be re-sent), and
thus can adversely impact speed-sensitive applications and services. Thus, high latency metrics

57

See RootMetrics, A simple premise for a sophisticated methodology, July 2015
http://www.rootmetrics.com/us/methodology.
17

 by themselves – even if reliable latency metrics could be obtained – cannot be evaluated without
understanding how they interact with the various other network metrics.58
Moreover, a myopic focus on only certain network metrics could harm customers. If the
Commission focuses on only certain under-the-hood network metrics, it would create incentives
for mobile broadband providers to adjust those metrics to meet certain thresholds, even if doing
so results in overall reduced network performance and customer experiences.
There are numerous other challenges to cobbling together reliable speed and other
metrics from the data sources identified in the Notice. For mobile wireless (and for wireline
services), there are many different ways to measure the proposed metrics, and it is not at all clear
how the Commission could use any of the proposed data sources identified in the Notice or
whether they could be combined in an apples-to-apples fashion. As just one example, speed and
latency metrics can vary dramatically depending on the endpoints of the connection and each of
the data sources appears to use different end points when taking these measurements. Thus, it is
not at all clear that cobbling together speed and other metrics from all of these sources is even
possible. It is clear, however, that any such attempt would require substantial Commission
resources, with no apparent benefits.

58

In addition, in the 2015 Broadband Progress Report, the Commission explained that it had
concerns about the use of “other factors, such as latency concerns” when it came to mobile
broadband. 2015 Broadband Progress Report ¶ 9. Earlier, in the Eighth Broadband Progress
Report, the Commission cited the lack of standards for measuring core characteristics such as
latency with respect to mobile broadband services. See Eighth Broadband Progress Report ¶ 3
(“[T]he Commission considered latency and capacity as core characteristics that affect what
consumers can do with their broadband service. The Commission identified potential standards
on latency and usage capacity with respect to fixed broadband services, but did not do so for
mobile services, and the latency and capacity of many mobile broadband services may not be
comparable to those of fixed broadband services”). None of these concerns have been resolved.
18

 CONCLUSION
For the foregoing reasons, the Commission should conclude that advanced
telecommunications capability is being deployed to all Americans in a reasonable and timely
fashion.
Respectfully submitted,

James P. Young
Christopher T. Shenk
Rishi P. Chhatwal
Marc A. Korman
Sidley Austin LLP
1501 K Street, N.W.
Washington, D.C. 20005
(202) 736-8000

/s/ Terri L. Hoskins
Terri L. Hoskins
Gary L. Phillips
David L. Lawson
AT&T Services, Inc.
1120 20th Street, N.W.
Washington, D.C. 20036
(202) 457-2055

Attorneys for AT&T

September 15, 2015

19