Before the FEDERAL COMMUNICATIONS COMMISSION Washington, D.C. 20544

Before the FEDERAL COMMUNICATIONS COMMISSION Washington, D.C. 20544
Before the
FEDERAL COMMUNICATIONS COMMISSION
Washington, D.C. 20544
In the Matter of
Framework for Broadband Internet
Service
Open Internet Rulemaking
)
)
)
)
)
)
)
GN Docket No. 10-127
GN Docket No. 14-28
COMMENTS OF VERIZON AND VERIZON WIRELESS
Of Counsel:
Michael E. Glover
William H. Johnson
Roy E. Litland
VERIZON
1320 North Courthouse Road
9th Floor
Arlington, VA 22201
(703) 351-3060
Attorneys for Verizon
and Verizon Wireless
Russell P. Hanser
WILKINSON BARKER KNAUER, LLP
2300 N St., NW
Suite 700
Washington, DC 20037
Helgi C. Walker
Kellam M. Conover*
GIBSON DUNN & CRUTCHER LLP
1050 Connecticut Ave., NW
Washington, DC 20036
*Admitted only in California; practicing under the
supervision of Principals of the Firm
July 15, 2014
TABLE OF CONTENTS
Page
I. II. III. TODAY’S OPEN INTERNET BENEFITS CONSUMERS, EDGE PROVIDERS,
AND BROADBAND PROVIDERS, AND A BALANCED APPROACH THAT
PROTECTS CONSUMERS AND COMPETITION WHILE ENCOURAGING
INNOVATION AND INVESTMENT WILL KEEP IT THAT WAY. ....................... 4 A. Verizon Strongly Supports an Open Internet and Has Strong Incentives to
Promote Internet Openness. ................................................................................ 5 B. Internet Openness Is Best Promoted Through A Flexible Framework Rather
than Prescriptive Rules....................................................................................... 10 FLEXIBILITY, INNOVATION, AND CONSUMER CHOICE SHOULD GUIDE
THE COMMISSION’S APPROACH, WHETHER OR NOT IT DECIDES TO
ADOPT FURTHER RULES.......................................................................................... 16 A.
Any Additional Transparency Requirements Must Be Reasonable and
Limited to the Disclosing ISP’s Own Network. ................................................ 21 B.
Any New No-Blocking Rule Should Ensure No Blocking of Content on the
Customer’s Selected Tier of Service But Should Allow Flexibility for
Additional Differentiated Services. ................................................................... 25 C.
Any Commercial Reasonableness Standard Should Allow Flexibility for ProConsumer Innovation and Experimentation. ................................................... 29 D.
Under the Proposed Commercial Reasonableness Rule, the Commission
Could Address Particular Practices That Harm Consumers or
Competition. ........................................................................................................ 36 ADDITIONAL REGULATION IS ESPECIALLY UNWARRANTED IN THE
MOBILE BROADBAND MARKETPLACE, WHICH PRESENTS UNIQUE
CIRCUMSTANCES AND IN WHICH AMERICA LEADS THE WORLD. .......... 38 A.
Unique Features Distinguish Mobile Broadband Service from Fixed
Broadband Service. ............................................................................................. 39 B.
Mobile Wireless Broadband’s Unique Features Warrant a Particularly
Light-Touch Regulatory Approach. .................................................................. 43 IV. BROADBAND RECLASSIFICATION WOULD BE A RADICAL AND RISKY
STEP THAT WOULD JEOPARDIZE NATIONAL GOALS AND VIOLATE THE
COMMUNICATIONS ACT. ......................................................................................... 46 V. THE COMMISSION SHOULD REFRAIN FROM IMPOSING ANY RULES ON
INTERNET INTERCONNECTION ARRANGEMENTS, SPECIALIZED
SERVICES, OR ENTERPRISE OFFERINGS............................................................ 69 VI. A. There Is No Reason to Apply Open Internet Rules to Peering, Transit, and
Other Internet Interconnection Arrangements................................................ 70 B. There Is No Reason to Apply Open Internet Rules to Specialized Services. . 76 C. There Is No Reason to Apply Open Internet Rules to Enterprise Services... 77 CONCLUSION ............................................................................................................... 78 Before the
FEDERAL COMMUNICATIONS COMMISSION
Washington, D.C. 20544
In the Matter of
Framework for Broadband Internet
Service
Open Internet Rulemaking
)
)
)
)
)
)
)
GN Docket No. 10-127
GN Docket No. 14-28
COMMENTS OF VERIZON AND VERIZON WIRELESS1
Maintaining an open, healthy and robust Internet is a paramount goal that benefits
consumers, edge providers, broadband providers, and countless others. Nothing about the
litigation that led to the remand of the Commission’s 2010 Open Internet Order, or the result of
that dispute, should distract observers from the core point that Verizon supports and relies upon a
robust and open Internet. Our customers demand it, and our business depends on it. We have
committed to our customers our support for the open Internet, and our broadband Internet access
services enable them to go where they want and do what they want online. We invest in worldclass broadband networks, such as our all-fiber FiOS network and our 4G LTE wireless network,
to keep pace with consumers’ demand and offer an ever-more-robust range of services. We also
are actively engaged in many other parts of the Internet ecosystem, including through our
Internet backbone networks, content delivery networks, over-the-top services, cloud-services,
and other innovative services that rely on the open Internet and enable a better Internet
experience. Each of these efforts depends on an open Internet and on the ability to reach
1
In addition to Verizon Wireless, the Verizon companies participating in this filing are the regulated, wholly
owned subsidiaries of Verizon Communications Inc. (collectively, “Verizon”)
1
consumers over the networks, software platforms and services of other providers. Verizon’s
various offerings create more choices for our end-user consumers and for those served by other
broadband providers. In each of these areas, we face competition from a large and growing
number of other providers, both traditional and non-traditional, while also increasingly
partnering with a wide range of players to develop and offer new services and offerings to
consumers. Broadband providers are responding to competition as one would expect and hope:
By deploying more and better facilities, expanding the speeds and capacities of their service
offerings, and offering consumers competitive prices. The light-touch approach pursued over the
last two decades by policymakers of both parties has been central to these successes, and it
continues to provide the foundation for innovation and investment that serves consumers well.
For the reasons explained below, further regulation of broadband providers’ behavior is
not needed at this time and would threaten the healthy dynamics fueling the growth and
continued improvement of the Internet and the many services it enables. Competition and
innovation are widespread throughout the Internet ecosystem, and ensure that the marketplace is
responsive to consumer demand. Indeed, end users have accessed the Internet literally billons of
times since the early 1990s, when Internet access service first became widely available, and yet
proponents of regulation can only point to a few isolated and dated incidents of alleged
problems. Existing legal protections, including the Commission’s transparency rule and
generally applicable antitrust and consumer protection laws, as well as multi-stakeholder groups,
already provide an effective backstop to prevent and address any future issues that could emerge.
Should the Commission decide nevertheless to adopt rules it must strive to pursue a balanced
approach that protects consumers and the open Internet while preserving the principles of
flexibility, innovation, and consumer choice that have marked its approach until now. A
2
continued focus on transparency, supported by the Commission’s existing rule, is central to such
an approach and should ensure that consumers have the information they need to understand
their service options and to make informed choices. The benefits of transparency to competition
on the Internet extend well beyond broadband providers’ services, as it is irrefutable that a wide
range of Internet companies beyond broadband providers principally shape the consumer
experience.
The Commission has also proposed to reinstate a no-blocking rule. Verizon has been
clear with our customers that we will not block their access to any content, applications, services
or devices based on their source. Any new no-blocking rule should be consistent with that
approach and ensure providers do not block any lawful Internet traffic within the customer’s
chosen Internet access service. But, consistent with the D.C. Circuit’s decision, the Commission
should allow flexibility for providers to negotiate differentiated arrangements or experiment with
different service models if they see a customer benefit in doing so, even as customers can
continue to go anywhere using their selected tier for Internet access service and edge providers
can rely on that service to reach their customers without the need for negotiating with broadband
providers for access to end-users.
Similarly, a commercial reasonableness standard, if appropriately applied, could allow
the flexibility for innovation that will benefit consumers while also providing a backstop that
would permit the Commission to address problems or market failures that arise and that harm
competition or consumers. Any commercial reasonableness standard should allow real
flexibility for new approaches that may better serve consumers, even as they continue to receive
traditional Internet services.
3
If the Commission proceeds with new regulations, it should limit such rules to fixed
broadband Internet access services, and should continue to recognize the unique circumstances
of the mobile marketplace – including the technical, competitive, and operational differences –
warranting a more restrained approach. This rapidly evolving sector requires particular
flexibility to serve consumers well, and high levels of innovation, investment and competition
resulting from the light-touch approach speak for themselves.
Finally, “reclassification” of broadband Internet access service as a Title II common
carriage telecommunications service would be a radical departure that would not achieve its
proponents’ stated goals and would only endanger the entire Internet ecosystem. The arcane
regulatory framework embodied in Title II was crafted for 19th Century railroad monopolies and
the early 20th Century one-wire telephone world. The price and service regulation inherent in
Title II have no place in today’s fast-paced and competitive Internet marketplace, and the threats
posed by this approach would not likely be confined to broadband providers but would spread
inevitably to other Internet sectors. Moreover, such an approach would be unlawful and, at a
minimum, would result in years of counterproductive uncertainty for the entire industry. In
contrast, a balanced framework will ensure that broadband providers act reasonably and would
protect against backsliding or bad acts that threaten consumers or competition, while preserving
flexibility for all providers to experiment with new approaches that could offer new choices and
benefit consumers and small Internet players alike.
I.
TODAY’S OPEN INTERNET BENEFITS CONSUMERS, EDGE PROVIDERS,
AND BROADBAND PROVIDERS, AND A BALANCED APPROACH THAT
PROTECTS CONSUMERS AND COMPETITION WHILE ENCOURAGING
INNOVATION AND INVESTMENT WILL KEEP IT THAT WAY.
A casual observer of the open Internet debate over recent months could be forgiven for
believing we are on the edge of a cataclysm, in which those seeking to preserve the Internet,
4
expand its reach, and extend its many capabilities square off against profit-hungry Internet
Service Providers (ISPs) aiming to extract ever-higher monopoly rents, undermine third-party
content providers, and squelch political debate. This view has been promoted by superficial
news reports, sensationalistic interest-group fund-raising appeals, and even late-night comedy
routines. But, as is so often the case, the truth is far less dramatic. Once the political cant and
commercial posturing are discarded, the core fact in this debate remains – namely, that ISPs such
as Verizon strongly favor, and benefit from, network openness. Verizon has invested billions of
dollars in businesses that rely on the open Internet, which our customers view as essential and
which is therefore a critical ingredient to our success as an ISP. Verizon has entered into adjacent
lines of business, offering content and related offerings that must transit the last-mile networks,
software platforms, search engines and services of other providers throughout the ecosystem.
From our perspective, the objective of this proceeding is not to subvert or evade principles of
Internet openness, but rather to ensure that the Internet remains open while also promoting the
innovation and customer-driven evolution that has made it the greatest communications platform
in modern history.
A. Verizon Strongly Supports an Open Internet and Has Strong Incentives to
Promote Internet Openness.
The suggestion that broadband providers have incentives to undercut Internet openness
fundamentally misunderstands the increasingly competitive and complex structure of the online
ecosystem. Consumers enjoy a dizzying array of services offered by a host of rival providers. In
today’s competitive and dynamic Internet marketplace, stakeholders often partner with one
another to provide certain services while competing in the provision of others – and the last-mile
5
access link is only one input among many.2 Entities that offer network capacity, online services,
applications, content, and devices simultaneously partner with and compete against one another
to offer broadband and related services.3 For example, Verizon Wireless partners with Google to
develop and advance the Android platform, while Verizon and Google compete in providing
cloud computing services. This kind of “mix-and-match” competition is fast collapsing
traditional distinctions between and among participants in the broadband arena.4
In this environment, companies such as Verizon have no incentive to abridge customers’
Internet freedoms or assign anyone to a slow lane. Verizon has invested tens of billions of
dollars to build our wired and wireless broadband networks, including more than $23 billion
spent to construct our fiber-optic FiOS network since 2004, and we continue to expend large
sums to improve and grow those networks.5 We must recoup these investments. The only way
to do so is to win and retain customers, and the only way to do that is to offer them the services
and features they demand – including access to the content and applications they want to use.
The suggestion that ISPs can benefit in the long term by limiting use of their networks and
funneling users to ISP-affiliated content is a politically motivated fantasy.
2
See, e.g., Jeffrey A. Eisenach, Broadband Competition in the Internet Ecosystem, AEI ECONOMICS
STUDIES, at 18 (Oct. 18, 2012), http://www.aei.org/files/2012/10/17/-broadband-competition-in-theinternetecosystem_164734199280.pdf (“Although it is certainly understandable that the modern
telecommunications intelligentsia would see broadband as the center of the Internet ecosystem … it is not. For
purposes of competition analysis, at least, broadband is a complement among complements, a module among
modules.”) (“BROADBAND COMPETITION”).
3
See generally Declaration of Andres V. Lerner, Competition in Broadband and “Internet Openness”, ¶ 6768 (July 15, 2014), attached hereto as Exhibit 2 (“Lerner Decl.”) (discussing platform competition).
4
See, e.g., BROADBAND COMPETITION at 18.
5
Verizon is targeting total capital investments of between $16.5 billion and $17 billion in 2014 alone. See
Verizon 2014 Investor Quarterly, First Quarter (Apr. 24, 2014), available at
http://www.verizon.com/investor/DocServlet?doc=vz_1q2014_bulletin.pdf. See also Lerner Decl., ¶ 41 (“When
Verizon launched FiOS in 2004, the company announced plans to spend about $23 billion over several years to build
out the necessary fiber infrastructure.”), ¶ 72 (“[A]nnual investment in U.S. wireless networks grew from $21 billion
in 2009 to $30 billion in 2012. In 2011, wireless broadband providers Verizon and AT&T were the top two
American firms in terms of capital expenditures.”).
6
Even if one assumed (wrongly) that there is some short-term gain to be had in this
manner, it would be vastly outweighed by the losses that would ensue when customers began to
discontinue service in favor of the ISP’s competitors. Verizon expends a significant amount to
build its network and attract and connect each FiOS customer we acquire, all with the hope that a
long-lasting relationship will allow us to recoup these investments and profit from each customer
over time. But, as illustrated by the churn that takes place in the industry, customers can and do
switch providers when their own provider makes them unhappy.6 In fact, “there is significant
customer switching in the broadband industry, particularly in recent years….A 2010
Commission study found that 36 percent of Internet users had switched broadband service
providers in the prior three years.”7 A short-term stratagem to obtain marginal increases in
revenues through “tolls” on edge providers or favoritism would backfire when the customer
inevitably switched to another ISP, resulting in the loss of thousands of dollars in potential
lifetime revenues from a satisfied customer. For example, if Verizon engaged in unpopular
practices that resulted in the loss of even just 2 percent of FiOS customers, that would equate to
more than $200 million in lost revenue per year. Thus, in our capacity as a broadband provider,
Verizon has strong reasons to promote Internet openness.
Our incentive to embrace openness is not, however, limited to the fact that our broadband
customers demand it. Verizon’s wireline footprint is limited, and we thus increasingly rely on
the Internet services of other broadband providers to serve our customers and to support the
strategic services in which we are investing for the future. To provide just a few examples,
Verizon (in partnership with Redbox) offers an over-the-top video service that is available over
6
See, e.g., Lerner Decl. ¶¶ 73-76.
7
Id. ¶ 46, citing Federal Communications Commission, Broadband Decisions: What Drives Consumers to
Switch – or Stick With – Their Broadband Internet Provider, at 3 (Dec. 2010),
https://apps.fcc.gov/edocs_public/attachmatch/DOC-303264A1.pdf.
7
any broadband connection in the country. We have focused heavily on the growing area of
cloud-based services through our Verizon Enterprise Solutions business. Our Verizon Digital
Media Services business provides a variety of services to facilitate online video and other data
processing and distribution across the Internet. With the recent purchase of EdgeCast, we now
operate a significant global content delivery network.8 In all these capacities, Verizon depends
upon the existence of an open Internet, and on the ability to reach consumers over the networks,
software platforms, search engines and services of other providers. Indeed, the vast majority of
Internet customers, both nationwide and globally, subscribe to broadband providers other than
Verizon, meaning that we must regularly send traffic over the networks of third-party ISPs, and
rely on other providers to direct and connect them to the content and services they desire.
Verizon would be deeply concerned if any player in the Internet ecosystem – another ISP, a
large content provider, a search engine, a major social network, an operating system provider, or
any other entity – were to keep end users from utilizing the Internet on their own terms or to
restrict the availability of Verizon’s services to those end users.
Put bluntly, Verizon’s own business model depends on market-driven Internet openness.
To this end, Verizon has publicly committed to customers that we will maintain an open network
for them. For example, our website promises users that they “can access and use the legal
content, applications, and services of your choice, regardless of their source” on “any of our
Internet access services, wireline or wireless,” “so long as they are legal and do not harm our
networks or the provision of Internet access service, facilitate theft of service, or harm other
users of the service.” Users likewise “can attach to the Verizon Wireless network any device
marketed by Verizon Wireless, or certified through the Verizon Wireless Open Development
8
See generally Verizon, What We Do, http://about.verizon.com/our-company/products-services.
8
program.” Verizon promises to provide customers “accurate and relevant information in plain
language about the characteristics and capabilities of our Internet access services so you can
make informed choices,” as well as “tools to keep track of … usage to avoid surprises….”9
Far from blocking user access to content and applications or advantaging deep-pocketed
Internet companies, Verizon has worked to promote third-party content and services, knowing
that such efforts are essential to winning and keeping customers. For example, we have
established state-of-the-art LTE Innovation Centers in Waltham, Massachusetts and San
Francisco, California that provide facilities, tools and assistance to innovators of all types in
developing the devices and applications of tomorrow.10 Verizon does this without charge to
these innovators, without claims to their intellectual property, and without any commitment that
they actually use any of their developments on Verizon’s networks. Our aim is to ensure that
these “garage innovators” have the resources and assistance they might otherwise lack, and
consequently will develop new offerings that will benefit our customers. We also operate a
venture capital business, Verizon Ventures, focused on partnering with and funding promising
Internet entrepreneurs. Similarly, we have established the Powerful Answers Award program to
encourage and reward those who develop new and promising approaches to the use of broadband
technology to address pressing needs for education, sustainability, health care, and
transportation.11 Verizon also recently became the first major communications service provider
9
Verizon’s Commitment to Our Broadband Internet Access Customers: Our Customers Get Everything the
Open Internet Has to Offer, http://responsibility.verizon.com/broadband-commitment/.
10
See Welcome to the Verizon Innovation Program, http://innovation.verizon.com/content/vic/en.html.
11
See Power Answers Award Program, http://www.verizon.com/powerfulanswers/award/.
9
to join the Open Invention Network to promote advances in open source software. This
commitment to openness has helped drive our industry-leading levels of customer satisfaction.12
B. Internet Openness Is Best Promoted Through A Flexible Framework Rather
than Prescriptive Rules.
In promoting Internet openness and the continued evolution of services in ways that
benefit consumers, the Commission should rely on a flexible approach that promotes
experimentation and innovation throughout the Internet environment while protecting consumers
and competition from actual harm, not on prescriptive one-size-fits-all rules. For more than 20
years, federal policy has reflected a bipartisan commitment to light-touch Internet regulation.13
As former FCC Chief Economist Michael L. Katz notes, “[f]lexibility has been critical to the
Internet’s success.”14 This flexible approach, which allows consumer demand to drive this
emerging and dynamic industry, has succeeded in unleashing extensive network investment and
facilities deployment.15 For example, in reliance on the Commission’s decisions to refrain from
applying utility-style regulation to new broadband networks and services, Verizon has invested
12
Verizon ranked first in Customer Satisfaction in the J.D. Power 2013 U.S. Residential Internet Service
Provider Satisfaction StudySM. See J.D. Power, Customer Satisfaction Is High among Internet Customers Who
Upgrade to Premium Speed Offerings To Boost Performance, (Sept. 26, 2013), http://www.jdpower.com/pressreleases/2013-us-residential-internet-service-provider-satisfaction-study.
13
In 1993, the Clinton Administration’s Information Infrastructure Task Force outlined guiding principles
and objectives on the national information infrastructure (“NII”), such as “promoting private sector investment,
through appropriate tax and regulatory policies.” The Task Force noted: “To increase the likelihood that the NII will
be both interactive and, to a large extent, user-driven, government must reform regulations and policies that may
inadvertently hamper the development of interactive applications.” National Information Infrastructure: Agenda for
Action (Sept. 15, 1993), http://clinton6.nara.gov/1993/09/1993-09-15-the-national-information-infrastructureagenda-for-action.html.
14
Declaration of Michael L. Katz, Protecting and Promoting Consumer Benefits Derived from the Internet,
¶ 3 (July 15, 2014) attached hereto as Exhibit 1(“Katz Decl.”); see also id. ¶¶ 7-15.
15
See, e.g., USTelecom, Broadband Investment, http://www.ustelecom.org/broadband-industry/broadbandindustry-stats/investment (noting that, as of 2012, the broadband industry “has invested more than $1.2 trillion
dollars since 1996” and that, in 2012, the combined annual capital expenditures by wireline, wireless, and cable
broadband providers reached $68 billion); Diana G. Carew and Michael Mandel, U.S. Investment Heroes of 2013:
The Companies Betting on America’s Future, PROGRESSIVE POLICY INSTITUTE, at 5 (Sept. 5, 2013),
http://www.progressivepolicy.org/wp-content/uploads/2013/09/2013.09-Carew-Mandel_US-Investment-Heroes-of2013.pdf (noting that Telecommunications and Cable companies invested $50.5 billion in capital expenditures in
2012, second only to energy at $56.1 billion).
10
more than $23 billion deploying our all-fiber FiOS network, and tens of billions more deploying
mobile broadband facilities. While Verizon consistently ranks among the top 2 or 3 capital
investors in the United States, its competitors take similar risks by investing heavily in
broadband infrastructure. Just since 2009, providers have spent nearly $250 billion to deploy
wired and wireless broadband networks.16 Today, more than 95 percent of the population has
access to fixed wireline services, generally from multiple providers. Similarly, the
Commission’s most recent Mobile Competition Report found that 82 percent of American
consumers can choose from at least four mobile broadband providers, while roughly 98 percent
can choose from at least two.17 And technology is advancing all the time: Wireline providers
are deploying fiber deeper into their networks,18 cable-based providers are touting their rollout of
DOCSIS 3.0 service,19 and mobile providers are racing to deploy 4G/LTE services20 (with
Verizon Wireless already offering 4G LTE to 97% of Americans) – even as they investigate new
protocols that could someday give rise to 5G services. As economist Andres Lerner concludes,
“[t]he current competitive environment, and massive historical and planned investments in
16
See Office of Science and Technology Policy & The National Economic Council, Four Years of
Broadband Growth, at 5 (June 2013), available at
http://www.whitehouse.gov/sites/default/files/broadband_report_final.pdf.
17
Implementation of Section 6002(b) of the Omnibus Budget Reconciliation Act of 1993, Sixteenth Report, 28
FCC Rcd 3700, at Table 9 (2013) (“Mobile Competition Report”).
18
AT&T Invests More Than $90 Million in Arkansas Wireless and Wireline Networks in First Half of 2013
(Aug. 9, 2013), http://www.att.com/gen/press-room?pid=24645&cdvn=news&newsarticleid=36847 (last visited July
10, 2014) (touting “Project Velocity IP” as “spurring the deployment of advanced fiber-optic connections to
thousands of Arkansas businesses”); AT&T – Upgraded DSL Winning 51% Against Cable, FAST NET NEWS (July
12, 2013 3:08 PM), http://www.dslprime.com/dslprime/42-d/4949-atat-upgraded-dsl-winning-51-against-cable.
19
DOCSIS 3.0 Market Thriving, ZACHS EQUITY RESEARCH (Dec. 17, 2013),
http://www.zacks.com/stock/news/117193/DOCSIS-30-Market-Thriving (last visited July 10, 2014) (noting that
“[m]ajor cable TV operators in the U.S. are aggressively deploying high-speed Wideband network based on the
DOCSIS 3.0 technology,” with a 14% spike in annual investment expected for 2014).
20
Paul Lambert, Successful LTE strategies: How to use LTE to build a compelling broadband strategy,
INFORMA TELECOMS & MEDIA, at 4 (July 2012), http://www.informatandm.com/wpcontent/uploads/2012/07/Successful-LTE-strategies-white-paper.pdf (noting that “LTE is the fastest-growing
cellular technology in history in terms of subscription-number growth”).
11
deploying broadband networks to meet consumer demands, have been achieved by relying on
competitive market forces, not through rigid regulation.”21 Moreover, investment is on the rise:
one recent report projects that wireline infrastructure spending on “last mile” and “backbone”
facilities will rise from $38.6 billion in 2013 to $42.9 billion by 2017,22 and studies project
capital expenditures by U.S. wireless broadband providers of around $35 billion a year from
2013-2017, up from around $25 billion a year for 2010-2012.23 In short, the nation’s decadeslong commitment to flexible Internet regulation has been a resounding success, and promises
more of the same if kept in place.
While the open Internet that consumers demand is here to stay, consumers deserve the
opportunity to benefit from new and additional choices in service offerings or business models
that have yet to be conceived.24 As Professor Katz notes, “consumers differ widely in the
importance that they attach to different characteristics of their Internet experience” and “[g]iven
[this] diversity of consumer preferences in the broadband user population, consumer welfare is
maximized when consumers are free to choose from among a range of different types of user
experiences.”25 Thus, the Commission should reject proposals to mandate a one-size-fits-all
approach at the expense of customer-focused innovation going forward.26 Those favoring the
21
Lerner Decl. ¶ 23.
22
Lerner Decl. ¶ 99 (citing Telecommunications Industry Association, TIA’s 2014-2017 ICT Market Review
and Forecast, at 3-46 (2014) (“TIA ICT Market Review”)).
23
See Lerner Decl. ¶ 100.
24
See id. ¶ 2 (“[The Commission should allow] providers to explore new service offerings or business
models, including differentiated arrangements with content and application providers that offer additional choices to
consumers, rather than locking in place a one-size-fits-all approach. Imposing ex ante regulation that restricts the
ability of broadband providers to experiment and deploy such business models, without a clear showing that such
provider conduct harms the competitive process, would distort market outcomes, reduce investment incentives, and
thereby harm consumers.”).
25
Katz Decl. ¶¶ 8-9.
26
Id. ¶ 3 (“Rules that restrict flexibility or take a one-size-fits-all approach risk distorting competition and
harming innovation, investment, and consumer welfare.”), see also id. ¶¶ 7-15.
12
highly prescriptive approach proclaim the benefits of the open Internet – benefits that no
stakeholder denies – but fail to demonstrate that this goal is better met by innovation-inhibiting
rules than by the existing flexible approach. Notwithstanding the trillions of bits traversing the
Internet every day, they can cite only a tiny handful of examples of arguable misconduct by
American broadband providers since the dawn of broadband service.27 They raise fears of “fast
lanes” and “slow lanes,” forgetting that large Internet incumbents such as Netflix and Google
already have implemented the equivalent of their own “fast lanes” that expedite the delivery of
their own traffic to consumers using in-house or third-party private networks; content-delivery
networks (CDNs); and even techniques such as Google’s implementation of the SPDY protocol
in ways that direct certain mobile traffic accessed through the Chrome browser to Google’s
proxy servers. Google’s approach directs more traffic to its network and thus allows more access
for Google to end-user data. In this environment, it is small upstarts as much as anyone that
could benefit from new business arrangements, which might help them compete against the
Internet’s content giants.
Nor is there reason to fear that the Internet will become a digital “dirt road.” To the
contrary, data tracked by the Commission and others shows that, in response to consumer
demands and competitive pressures, the capabilities of broadband networks have continually
increased, as domestic broadband providers outpace most international counterparts in investing
27
Preserving the Open Internet; Broadband Industry Practices, Report and Order, 25 FCC Rcd 17905, ¶ 35
(2010) (“Open Internet Order”) (citing to settlement of alleged blocking by Madison River Communications, LLC
and enforcement action against Comcast Corporation concerning subscriber uploads of peer-to-peer (P2P) content).
13
in more advanced broadband infrastructure.28 For example, total capital expenditures by U.S.
wireline broadband providers in 2012 equaled $25 billion.29 Per capita broadband investment in
this country is roughly double that of European counterparts.30 As a result, the speeds
associated with today’s “best efforts” Internet service are double those of just four years ago,
when the Commission last considered open Internet issues.31 This consistent, aggressive growth
in broadband speeds ensures that the best-effort Internet will remain the principal means by
which Americans access an ever-growing range of diverse content at higher speeds and with less
latency or jitter than before.
28
See, e.g., 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, Eighth
Broadband Progress Report, 27 FCC Rcd 10342, 10347 ¶ 6 (2012) (“Recent trends show providers offering much
higher speeds” through advances in both wireline and wireless technologies); see also Office of Engineering and
Technology & Consumer and Governmental Affairs Bureau, 2012 Measuring Broadband America : A Report on
Consumer Wireline Broadband Performance in the U.S., FEDERAL COMMUNICATIONS COMMISSION, at 4 (July
2012), http://transition.fcc.gov/cgb/measuringbroadbandreport/2012/Measuring-Broadband-America.pdf (finding
“striking across-the-board improvements on key metrics underlying user performance”); International Comparison
Requirements Pursuant to the Broadband Data Improvement Act; International Broadband Data Report, Third
Report, 27 FCC Rcd 9884 (2012). See also, e.g., Joan Engebretson, Akamai: Average U.S. Broadband Speed
Climbs to 9.8 Mbps, TELECOMPETITOR (Jan. 29, 2014, 1:00 AM), http://www.telecompetitor.com/akamai-average-us-broadband-speed-climbs-9-8-mbps/; Andrew Burger, Report: Average U.S. Broadband Prices Are Below World
Average of $76.61, TELECOMPETITOR (April 30, 2014, 7:15 PM), http://www.telecompetitor.com/report-average-u-sbroadband-prices-are-below-world-average-of-76-61/.
29
See Lerner Decl. ¶ 100. For U.S. wireline industry providers overall one recent report projects that
infrastructure spending on “last mile” and “backbone” facilities, will rise from $38.6 billion in 2013 to $42.9 billion
by 2017. Id. ¶ 99, citing TIA ICT Market Review at 3-46.
30
Christopher S. Yoo, U.S. vs. European Broadband Deployment: What Do the Data Say?, Center for
Technology, Innovation and Competition (June 2014), https://www.law.upenn.edu/live/files/3353-us-vs-europeanbroadband-deployment-summary (last visited July 15, 2014) (“Data analysis indicates that as of the end of 2012, the
U.S. approach promoted broadband investment, while the European approach had the opposite effect ($562 of
broadband investment per household in the U.S. vs. $244 per household in Europe.”).
31
Compare Akamai, Akamai State of the Internet Report Spotlights Latest Global Speed and Attack Trends
from Fixed and Mobile Internet Connections (Jan. 24, 2011),
http://www.akamai.com/html/about/press/releases/2011/press_012411.html (overall average connection speed for
the U.S. as a whole in the third quarter of 2010 was 5.0 Mbps and overall average peak connection speed was 20
Mbps.) with Akamai’s State of the Internet Q4 2013 Report, at 23 (2014),
http://www.akamai.com/dl/akamai/akamai-soti-q413.pdf (average connection speed in the United States were 10.0
Mbps and average peak connection speeds were 43.7 Mbps.).
14
In contrast to the long-standing light-touch approach that has fostered these successes, a
regime centered on inflexible rules would undercut the innovation and investment that
characterize today’s Internet, making regulatory bodies in Washington, D.C. the new epicenter of
decision-making for all things related to the Internet. The negative consequences would be
especially harmful to consumers and competition if applied in the context of mobile wireless
services, which, as explained further below, are still developing, are provisioned in an intensely
innovative and competitive environment, and pose unique technical challenges not associated
with fixed networks.32
In particular, as addressed in Section IV below, the Commission should reject arguments
by those asking it to turn the clock back by more than a century and “reclassify” broadband
Internet services as including a distinct “telecommunications” component subject to common
carrier regulation. This radical departure from well-established precedent would be contrary to
the facts and the law.33 Moreover, the ultra-regulatory “Title II” approach would raise the
specter of rate regulation, entry and exit regulation, and a broad array of other requirements,
prompting investors to withdraw capital from the communications sector (as they did when the
idea was first considered in 2010).34 Furthermore, because all Internet services involve some
form of transmission, an approach severing an information service’s transmission component
from its processing capabilities would also implicate a wide range of information service
providers. For example, large Internet incumbents like Amazon and Google provide much of
32
See Katz Decl. ¶¶ 3, 62-63.
33
See generally infra Section IV.E.
34
See generally Cecilia Kang, A look at how the FCC’s move can affect stocks, WASHINGTON POST TECH
BLOG (May 7, 2010, 11:07 AM),
http://voices.washingtonpost.com/posttech/2010/05/a_look_at_how_the_fccs_move_ca.html (showing that, the day
after Chairman Genachowski floated his “Third Way” reclassification proposal, “shares of cable and telecom stocks
were treading 300 to 400 basis points lower than the overall market”).
15
their own transmission, and operate some of the largest networks for delivering that traffic.35 All
of the Internet-based services they offer consumers, including search engines, social networks,
and messaging applications, also incorporate transmission. All of these offerings could be swept
up into Title II’s reach under any reclassification. Ironically, even as it imposed all these harms,
reclassification would not even preclude the differentiation of service that its proponents seek to
ban, because Title II expressly recognizes that reasonable discrimination is lawful and has long
permitted many of the practices that Title II proponents criticize. Thus, the application of Title II
requirements to broadband providers would amount to regulation for the sake of regulation,
strapping a straightjacket onto this competitive and dynamic sector.
II.
FLEXIBILITY, INNOVATION, AND CONSUMER CHOICE SHOULD GUIDE
THE COMMISSION’S APPROACH, WHETHER OR NOT IT DECIDES TO
ADOPT FURTHER RULES.
As evidenced by the absence of problems in the broadband marketplace to date, a regime
that relies in the first instance on informed consumer choice protects broadband consumers
through the very same mechanisms that protect consumers in myriad other industries: The
disciplining force exerted by informed consumers making choices regarding their own best
interests, backed by flexible legal protections to address bad actors in the event of market
breakdowns.36 “Well-informed consumers are the best judges of their own preferences.”37 This
35
In addition to operating its own CDN, Google is in the process of constructing a substantial last-mile
presence through “Google Fiber,” which already serves several cities and may soon serve dozens more. See, e.g.,
Lerner Decl. ¶ 33.
36
See, e.g., Lerner Decl. ¶ 47 (“[B]roadband consumers tend to be well informed. In the broadband industry,
marketing and advertising raise consumer awareness of differences between broadband providers’ product offerings,
including speed, pricing, and data caps. Broadband providers’ advertising and marketing efforts provide useful
information to consumers, by comparing a given providers’ offerings with those of competitors. Additionally,
industry groups and publications (e.g., Consumer Reports, J.D. Power and Associates) provide surveys and
comparisons of broadband provider offerings, and report user satisfaction with various providers, allowing
consumers to make informed decisions when choosing providers.”).
37
Katz Decl. ¶ 9.
16
approach directs investment to the services that consumers demand, and thus will enhance the
value of those services more effectively than regulation.
While transparency and informed consumer choice will continue to protect Internet
openness and deter harmful practices on the Internet, additional safeguards do and should exist to
address any new issues that may arise and to address unreasonable practices that harm
competition or consumers. Indeed, the Internet ecosystem has a long and successful history of
addressing problems as they arise even without regulatory involvement through multistakeholder processes. Groups such as the Internet Engineering Task Force, the Internet Society
and others have played a valuable role in this regard and have helped to guide the evolution of
Internet technology and to address emerging issues. The Broadband Internet Technical Advisory
Group (BITAG) is a more recent example that provides a forum for airing concerns and
addressing technical issues while avoiding the downsides of a more regulatory approach.
BITAG was created to “discuss and opine on technical issues pertaining to the operation of the
Internet, as a means of bringing transparency and clarity to network management processes as
well as the interaction among networks, applications, devices and content.”38 Aside from such
multi-stakeholder groups, generally applicable antitrust and consumer protection laws also
remain available to address many forms of alleged misconduct.39 For example, consumer
protection laws, including those administered by the Federal Trade Commission, already guard
against fraud, deception, and similar practices.40
38
See BITAG, BITAG Organization, BITAG History, available at
http://www.bitag.org/bitag_organization.php?action=history.
39
See, e.g., Katz Decl. ¶ 29.
40
Thus, no additional rules are needed to prohibit, for example, a broadband provider from affirmatively
misleading consumers through a false statement about a material term related to its service or failing to provide
access to all lawful content and applications if that is what the service provider has promised.
17
Taking into account these existing protections, the Commission can continue to play an
important role in addressing unreasonable conduct should it arise. Rather than pursuing a
traditional, prescriptive approach to regulation that would be ill-fitting and counterproductive for
the dynamic Internet marketplace, however, the Commission could act as a “backstop,” stepping
in when needed to address unreasonable actions that harm competition or consumers. This type
of case-by-case, enforcement-based approach should provide the flexibility necessary to
encourage experimentation and innovation, while still allowing the Commission to intervene and
address practices that demonstrably harm consumers or competition.41
This proposed framework also fosters the flexibility that the NPRM recognizes is needed
to “permit broadband providers to engage in individualized practices” that can expand
consumers’ options.42 Flexibility to offer new products and services is key to success in the fastpaced marketplace of broadband and related Internet-based services.43 For example,
experimentation with new competitive offerings – such as T-Mobile’s recently announced
“Music Freedom” plan, which offers users unlimited streaming of music without incurring any
additional usage-based charges44– creates new choices for consumers and provide additional
dimensions on which providers can compete. It is instructive that T-Mobile’s experiment with
an approach that, on its face, gives consumers new benefits and potential savings was met with
41
See Katz Decl. ¶ 26 (describing benefits of case-by-case approach).
42
Protecting and Promoting the Open Internet, Notice of Proposed Rulemaking, GN Docket No. 14-28, FCC
14-61, ¶ 111 (May 15, 2014) (“NPRM”). Thus the NPRM focuses on “requiring broadband providers to use
‘commercially reasonable’ practices in the provision of broadband Internet access service,” an approach that is
“more flexible than the vacated 2010 non-discrimination rule” and would “permit broadband providers to serve
customers and carry traffic on an individually negotiated basis.” Id. ¶ 116.
43
See Katz Decl. ¶ 13.
44
See http://www.t-mobile.com/offer/free-music-streaming.html.
18
knee-jerk hostility from some net neutrality proponents.45 Flexibility to test the appeal of such
offerings in the competitive marketplace does not undercut Internet openness.46 Regardless of
what innovations may ultimately come to pass or be embraced by consumers, today’s consumers
have made clear that they expect and demand access to whatever lawful content, applications,
and services they choose, whenever they want it. And Verizon and other broadband providers
have committed to meeting these consumer demands and expectations.
To the extent the Commission does adopt new rules governing the broadband
marketplace; moreover, it should ensure that those rules apply in an even-handed way that
promotes innovation across the ecosystem. Large Internet incumbents such as Google,47
Netflix,48 and Amazon have undeniable power to affect the consumer experience online and
Internet openness, and the reach of these companies often dwarfs that of particular ISPs. For
example, Google exercises pervasive influence over consumers’ Internet usage: It controls more
45
See, e.g., John Eggerton, T-Mobile Criticized for Online ‘Music Freedom’ Move,
BROADCASTINGCABLE.COM (June 20, 2014 12:19 PM), http://www.broadcastingcable.com/news/washington/tmobile-criticized-online-music-freedom-move/131915.
46
See Katz Decl. ¶¶ 7-15.
47
Trouncing the market share of any provider in the national broadband services marketplace, Google’s share
of the search market is nearly 68 percent. See Ashley Zeckman, Google Search Engine Market Share Nears 68%,
SEARCH ENGINE WATCH (May 20, 2014), http://searchenginewatch.com/article/2345837/Google-Search-EngineMarket-Share-Nears-68 (last visited July 10, 2014). Google’s practices related to search and other activities have an
undeniable impact on consumers’ experience on the Internet.
48
Netflix now boasts more than 48 million subscribers. See Brian Solomon, Green Is The New Black:
Netflix's Reed Hastings Joins Billionaire Ranks, FORBES INVESTING BLOG (Jun. 9, 2014, 2:38 PM),
http://www.forbes.com/sites/briansolomon/2014/06/09/netflix-ceo-reed-hastings-joins-billionaire-ranks/. That is
nearly 10 times the number of current Verizon FiOS subscribers, and more than five times the number of current
Verizon wireline broadband connections. See Verizon Reports Fifth Consecutive Quarter of Double-Digit
Operating Income and Earnings Growth: 1Q 2014 Highlights (Apr. 24, 2014),
http://www.verizon.com/investor/news_verizon_reports_fifth_consecutive_quarter_of_doubledigit_operating_inco
me_and_earnings_growth_042420.htm. Netflix’s subscribership is also more than four times the number of Uverse high-speed customers reported by AT&T. See AT&T Reports Strong Results in First Quarter while Investing
in Growth Transformation, http://about.att.com/story/att_first_quarter_earnings_2014.html (Apr. 22, 2014)
(reporting 11 million U-verse high-speed Internet customers).
19
than two-thirds of the American search market,49 exercising tremendous discretion over the
ordering of search results and thus over which websites Americans visit. It likewise controls the
distribution of content over its extensive and quickly expanding network plant. And by directing
more traffic to its own proxy servers as part of Google’s implementation of the SPDY protocol,
it is promoting further customer reliance on its network, and capitalizing on that reliance to
harvest more information regarding its users’ behavior.
Similarly, Netflix boasts of its power in the Internet ecosystem: Its website states that
“[t]he more successful Netflix becomes, the more important we are to the ISPs’ subscribers,” and
that “[i]t is natural for ISPs to worry that we may try to charge them in the future like linear TV
networks charge MVPDs.”50 Indeed, even aside from the popularity of Netflix with millions of
Internet users, the volume of Netflix’s traffic – more than one-third of all U.S. Internet traffic
during peak times, according to one recent study – gives Netflix significant power to affect how
the Internet operates and the consumers’ Internet experience.
Amazon exercises similar control over large swaths of the consumer Internet experience,
given the central role of its site in online commerce, the popularity of its consumer devices such
as the Kindle, and its cloud-based services. Amazon has even begun to limit customers’ ability
to purchase content from publishers that do not agree to its ebook pricing mechanism.51
Under these conditions, it would defy logic to focus only on an ISP’s alleged “incentives
and abilities,” but to ignore the roles played by other Internet companies. To be effective and to
avoid distorting competition, efforts to promote openness must apply to all relevant segments of
49
See Todd Bishop, Google’s market share climbs in latest U.S. search stats, GEEKWIRE (Jan. 15, 2014, 2:14
PM), http://www.geekwire.com/2014/googles-market-share-climbs-latest-u-s-search-stats/.
50
See Netflix Long-Term View, http://ir.netflix.com/long-term-view.cfm (last updated Apr. 21, 2014).
51
See, e.g., Carolyn Kellogg, Amazon and Hachette: The dispute in 13 easy steps, LOS ANGELES TIMES (June
3, 2014, 4:03 PM), http://www.latimes.com/books/jacketcopy/la-et-jc-amazon-and-hachette-explained-20140602story.html#page=1.
20
the market. This is not to say that extensive regulation is needed in any corner of the Internet
ecosystem at this point—there is no evidence that it is—but instead that whatever regulatory
backstop is deemed appropriate should account for the power exercised by all players in the
Internet ecosystem.
A. Any Additional Transparency Requirements Must Be Reasonable and Limited
to the Disclosing ISP’s Own Network.
In the competitive broadband marketplace, the Commission’s transparency requirement
can be and should be the linchpin in its open Internet policy framework. Appropriately framed
disclosures promote market interactions by ensuring that consumers understand the strengths and
weaknesses of competing offerings and can choose the ones best suited to their needs.52 Thus,
transparency enables consumer choice, a strong force for ensuring that the Internet remains
open.53 When consumers have useful information regarding the services available to them and
can make choices based on that information, additional prescriptive rules are unnecessary. The
Commission should continue to leverage transparency and consumer choice as means of
identifying and preventing practices that could harm consumers or competition.
Verizon provides customers with extensive information regarding our service offerings
and network management practices, and other ISPs appear to do the same. Indeed, given the
power of consumer demand in the broadband marketplace,54 providers have every reason to
disclose relevant information about their efforts to promote an open Internet (not to mention the
other capabilities of their advanced networks) and thereby win customers, absent any
52
See Katz Decl. ¶ 20 (“Transparency regarding available service offerings and network management
practices can promote competition and consumer welfare by allowing consumers to make better-informed
choices.”).
53
See, e.g., Katz Decl. ¶ 9 (“Well-informed consumers are the best judges of their own preferences. Given
the diversity of consumer preferences in the broadband user population, consumer welfare is maximized when
consumers are free to choose from among a range of different types of user experiences.”).
54
See supra Section I.B.
21
government mandates.55 However, the Commission must ensure that its disclosure mandates do
not become unduly complex and burdensome, such that they impose substantial costs without
providing any commensurate benefits.
The current disclosure rule is designed to elicit a broad range of detailed information
relevant to consumers, edge providers, and other stakeholders. The 2010 Open Internet Order
and the subsequent “advisory guidance” issued by the Office of General Counsel and the
Enforcement Bureau list numerous data points that broadband providers must disclose to
customers and edge providers, including the following:

The provider’s network practices, including congestion management,
application-specific behavior, device attachment rules, security measures, the
types of traffic subject to such practices, the purposes served by them, the
practices’ effects on end users’ experience, and the criteria used to effectuate
such practices;

The service’s performance characteristics, including a general description of
system performance (such as speed and latency); its suitability for real-time
applications; the effects of specialized services on available capacity; usage limits
and the consequences of exceeding them; and whether and why the provider
blocks or rate-controls specific protocols or protocol ports, modifies protocol
fields in ways not prescribed by the protocol standard, or otherwise inhibits or
favors certain applications or classes of applications;

Commercial terms associated with the service, including pricing, termination
fees, privacy policies, and redress options.56
In several places, the NPRM suggests more disclosures are needed, but then describes
information that is already covered by the existing rule. The NPRM’s discussion of customer
disclosures indicates that some customers need information concerning such topics as their
services’ speeds, the charges that will be applied, the sources of network congestion, or limits
55
See, e.g., Lerner Decl. ¶ 49; see also id. at n.53.
56
See generally Open Internet Order ¶ 55; FCC Enforcement Bureau and Office of General Counsel Issue
Advisory Guidance for Compliance With Open Internet Transparency Rule, Public Notice, 26 FCC Rcd 9411
(2011).
22
imposed as a result of “excessive use.”57 Elsewhere, it asks whether broadband providers should
be required to disclose “data caps.”58 Each of these examples relates to information that
providers must already disclose under the existing rule. If they are not doing so, the proper
response is to ensure compliance with the existing mandates, not to add unnecessary detail to
them or to create new mandates.
The NPRM also asks whether broadband providers should be required to disclose
“effective download speeds, upload speeds, latency, and packet loss.”59 While providers already
disclose much of this information, they necessarily do so in ways that account for the many
factors affecting performance, many of which are outside of the control of the provider.
Mandatory disclosures at the level of specificity suggested in the NPRM, however, could limit
provider flexibility, resulting in disclosures that were confusing, misleading, or both. On the
“best efforts” Internet, a particular service’s speed, latency, jitter, and other core characteristics
will vary based on a large array of factors, including the number of users accessing shared
network resources and the type of content being accessed. There would be no way to report this
information concisely, and such disclosures would be meaningless to all but the most technically
sophisticated customers. Worse, given the NPRM’s assumption that “consumers have difficulty
understanding commonly used terms associated with the provision of broadband services,”60 such
disclosures would simply deter consumers from reviewing the information provided, eviscerating
the rule’s effectiveness.61
57
See NPRM, ¶ 69.
58
Id. ¶ 72.
59
Id. See also id. ¶ 73 (asking whether ISPs should be required to report on “packet loss, packet corruption,
latency, and jitter in addition to upstream and downstream speed”).
60
Id. ¶ 68.
61
See Katz Decl. ¶ 22 (“There is a risk that some of the information the Commission seeks to make available
may be unintelligible to most end users.”).
23
Likewise, a rule mandating disclosure of even more information – whether to customers,
edge providers, or other stakeholders – could limit broadband providers’ ability to implement
innovative new network-management tools and security protocols, because it would require them
first to describe any new practices to consumers and edge providers – and, by extension, to
competitors, hackers, spammers, and others wishing to exploit the network’s weaknesses.
Broadband providers will naturally be averse to disclosing information that is competitively
sensitive or that exposes their networks and customers to external threats. Thus, excessive
disclosure requirements could have the perverse effect of retarding the network’s evolution by
deterring the deployment of new tools designed to help consumers.
Finally, if it chooses to impose additional transparency requirements, the Commission
must ensure that obligations falling on broadband service providers relate only to the last-mile
networks that those providers operate, and not to the networks of third parties. The NPRM
“tentatively conclude[s] that [it] should require that broadband providers disclose meaningful
information regarding the source, location, timing, speed, packet loss, and duration of network
congestion.”62 At the same time, the Commission acknowledges that “sources of congestion that
impact end users may originate beyond the broadband provider’s network or in the exchange of
traffic between that network and others.”63 This is true: Content providers and the backbone
networks on which they (and their own last-mile providers) rely make many choices regarding
how to route traffic,64 and those choices affect congestion and the speed at which customers are
62
NPRM ¶ 83.
63
Id. ¶ 82.
64
See infra Section V.A.
24
able to access content.65
Last-mile broadband providers cannot control what happens beyond
their networks, or how content providers choose to reach ISPs’ networks. Moreover, if an ISP’s
end user accesses content that has already been degraded by congestion before reaching the ISP’s
network (for example, on the network of a transit provider, a peering partner, or the last-mile
network serving the content provider), the ISP will have no way of determining the source of that
degradation.66 Under these circumstances, disclosures regarding congestion could create the
misleading impression that the ISP caused congestion, even when other network providers and
their choices are responsible for the issue.67 Thus, any new disclosure requirements addressing
the sources of congestion must be appropriately cabined to ensure that ISPs are only responsible
for reporting on the sources of congestion on their own last-mile networks.
B. Any New No-Blocking Rule Should Ensure No Blocking of Content on the
Customer’s Selected Tier of Service But Should Allow Flexibility for Additional
Differentiated Services.
Verizon is fully committed to delivering the content and applications that our customers
demand regardless of the identity of the edge provider involved, and will not block any lawful
Internet traffic as customers use their selected tier of broadband Internet access service. As
explained above, Verizon has publicly committed to our customers that, on any of our Internet
65
See Katz Decl. ¶ 23 (“the Commission’s transparency rule should be properly cabined to ensure that
broadband Internet access providers are responsible for reporting only on the sources of congestion for which they
are responsible, while recognizing that the decisions and practices of others in the Internet ecosystem likewise affect
performance and end users’ experiences”).
66
For example, a researcher studying Netflix’s speed index rankings recently concluded that the ranking
primarily reflected a range of factors unrelated to particular broadband providers’ networks, including a number of
factors within Netflix’s control. See Peter Sevcik, How the Netflix ISP Speed Index Documents Netflix Congestion
Problems, NETFORECAST (June 2014), http://www.netforecast.com/wpcontent/uploads/2014/06/NFR5117_How_the_Netflix_ISP_Speed_Index_Documents_Netflix_Congestion_Problem
s.pdf.
67
See Katz Decl. ¶ 23 (“To require broadband Internet access providers to disclose performance information
unrelated to their own facilities or services is potentially misleading to consumers.”). Nor can ISPs be expected to
identify alternative sources of congestion – any such “disclosure” would either reflect pure guesswork or require the
ISP itself to investigate the numerous networks that might be causing the problem. This would, of course, be
untenable.
25
access services, users “can access and use the legal content, applications, and services of [their]
choice, regardless of their source,” and can connect their “choice of technically compatible
devices,” so long as the content or devices at issue “do not harm our networks or the provision of
Internet access service, facilitate theft of service, or harm other users of the service.”68
Consistent with these promises, Verizon does not block or degrade any content or application
within our Internet access services on the basis of the views expressed, the source of the traffic or
identity of the content provider, or the extent (if any) to which the content or application
competes with a Verizon service offering. Verizon’s customers demand and desire open Internet
services, and value Verizon’s services precisely because they afford access to all the lawful
content and applications the Internet makes available. A policy of impeding access to services
customers wish to access would only push those customers to other providers. Other ISPs face
the same incentives. Under these circumstances – in which users demand access to all lawful
content and ISPs are committed to fulfilling customers’ needs – there is no need for a
prescriptive no-blocking rule.
To the extent that the Commission nonetheless decides to adopt a no-blocking rule, the
best regulatory framework would be a rule that prohibits providers from blocking or degrading
traffic within the customer’s chosen level of best efforts Internet access service based on who
sent it: if a customer buys a generic Internet access service offering download speeds of up to 50
Mbps, then Internet traffic from the provider’s affiliates or partners cannot be specially
provisioned to travel at 50 Mbps while traffic from other edge providers travels only at 25 Mbps.
Instead, all Internet traffic, regardless of who sent it, should get the same best efforts treatment
68
See Verizon’s Commitment to Our Broadband Internet Access Customers, supra n.10.
26
within the customer’s Internet service.69 With this approach, customers are assured that they can
use their best efforts Internet service to go where they want and do what they want online, and
edge providers have the comfort of knowing that they will not be disadvantaged within the last
mile and will have access to their customers.70
As today, this approach would mean that there would be no necessity for any edge
provider to seek out and enter into individual arrangements with broadband providers in order to
reach their customers. At the same time, and consistent with the approach discussed by the D.C.
Circuit in Verizon, edge providers, content providers, or others would have the flexibility to
negotiate differentiated arrangements with broadband providers if they believe they can provide
a service customers may want.71 Thus, all of the benefits of the traditional approach would
remain intact, while consumers potentially could gain additional choices and edge providers
could pursue alternative ways to distinguish themselves in the competitive marketplace.
This approach also would best account for evolving levels of service. Whereas a
Commission-imposed standard would either be static or require constant revision, a framework
based on ensuring that customers can use their selected tier of best efforts Internet access service
to go where they want and do what they want online will evolve on its own.72
Broadband
speeds have consistently increased, and continue to do so at a rapid pace, as consumers demand.
For example, 51 percent of Verizon FiOS Internet customers as of March 2014 subscribed to
69
Thus, at times when that customer’s Internet service is providing 50 Mbps over her connection, any
Internet content that the subscriber selects should get the full benefit of that speed over that last mile connection.
And if for whatever reason only 40 Mbps is available for the customer’s Internet service at a particular point in time
based on the myriad factors that can affect performance, again, that would be true regardless of the source of the
Internet traffic the subscriber seeks.
70
See Katz Decl. ¶ 32 (“A carefully crafted rule that ensures that traffic will not be blocked or degraded over
an end user’s best-effort Internet access service would provide assurances that end users could access the content
and applications that they desire and that edge providers would continue to have a path to reach end users.”).
71
Verizon v. FCC, 740 F.3d 623, 657 (2014).
72
See Katz Decl. ¶ 33.
27
FiOS Quantum, which offers speeds of 50 Mbps downstream/25 Mbps upstream or faster.73
Mobile broadband speeds continue to increase as well: In May, Verizon announced the rollout
of XLTE, which “delivers faster peak data speeds and a minimum of double the bandwidth to 4G
LTE customers in high traffic areas in markets nationwide where AWS spectrum has been
activated.”74 The approach suggested above would reflect these changes as they occur, rather
than leave outdated standards in place while the Commission repeatedly works to update them.
This “consumer choice” framework for a no- blocking rule would be superior to one
focused on quantitative performance criteria.75 Efforts to enumerate “specific technical
parameters,” such as “specific speeds of service” or specific outcomes that must be avoided (e.g.,
high jitter rates), would be doomed to failure: The capabilities of broadband networks are
constantly evolving, as are the needs of consumers. As the NPRM acknowledges, “a specific
technical definition of minimum access could become outdated as available broadband network
technologies change and available broadband speeds improve.”76 Moreover, different platforms
present different strengths and limitations, further complicating any effort to establish
performance benchmarks. Even if the Commission could identify relevant technical criteria,
adoption of such criteria would impose a one-size-fits-all solution on a diverse marketplace.
An approach that forbade ISPs from blocking any lawful Internet traffic as customers use
their selected Internet access service would solve these problems: Evolving capabilities and
demands would be reflected by the market and by evolving service arrangements, such that the
73
See TV Redefined: Verizon Brings FiOS Quantum TV Service to Maryland, Virginia and D.C. (May 20,
2014), http://newscenter.verizon.com/corporate/news-articles/2014/05-20-fios-quantum-tv-md-va-dc/.
74
See Verizon, XLTE: America’s Best Network Gets Even Better (May 19, 2014),
http://www.verizonwireless.com/news/article/2014/05/verizon-wireless-xlte.html.
75
See NPRM ¶ 103.
76
Id.
28
best efforts level of service purchased by the consumer would itself evolve over time.
Differences between platforms would be reflected in the services purchased by customers.
Customers would not be subject to one-size-fits-all standards, but rather would benefit from
knowing that their traffic would not be blocked.
This approach to a no-blocking rule would also be superior to the NPRM’s “reasonable
person” standard.77 The “reasonable person” standard would substitute the Commission’s view
of what consumers expect for the choices made by actual consumers. This result would reduce
consumer welfare, because different customers have different preferences. “It is not possible to
predict consumer demand, or the types of business models that undoubtedly will emerge. As a
result, a one-size-fits-all approach is likely to deny consumers choices that they may otherwise
find attractive.”78 There is no reason for the Commission to try to divine what a hypothetical
“reasonable consumer” might expect in terms of service or mandate such a one-size-fits-all
approach when consumers are making actual, observable choices every day about the levels of
service they require, and different “reasonable” consumers use the Internet in very different
ways.
C. Any Commercial Reasonableness Standard Should Allow Flexibility for ProConsumer Innovation and Experimentation.
An approach that relies on informed consumer choice subject to protections against
blocking would be ineffective if the Commission nevertheless prohibited broadband providers
from providing the services that consumers demand. Any rules that the Commission adopts to
regulate broadband providers’ services, therefore, should allow ample flexibility to foster
77
See id. ¶ 104.
78
Lerner Decl. ¶ 26; see also Katz Decl. ¶¶ 8-15.
29
consumer choice and drive pro-consumer innovation.79 As Professor Katz notes: “Because no
one has the ability to predict what will be the best network management practices and pricing
and service models in the future, it is important that the Commission’s rule be flexible and
remain focused on addressing only those practices that demonstrably harm competition. A caseby-case (or rule-of-reason) approach can offer that flexibility.”80 In this respect, the
Commission’s balanced proposal for a commercial reasonableness standard is on the mark. The
Commission strikes the right balance by proposing to allow flexibility for broadband providers to
develop new alternatives that consumers and edge providers might want, while still preserving
the ability to address specific practices that are determined to be anticompetitive or harm
consumers.81
In order to allow experimentation with additional service offerings or business models
and preserve individualized negotiations with other players in the Internet ecosystem, any
commercial reasonableness rule must be truly open to tailored arrangements that help fixed
broadband providers adapt their services to consumer choices, so long as there is no harm to
competition or consumers. This kind of experimentation with individualized agreements offers
the potential to initiate new “virtuous circles” that can benefit consumers, edge providers, and
broadband providers alike. For example, many such arrangements could reduce consumer costs,
whether through tiered pricing, ad-supported services, 800-type “sponsored data” services, two-
79
See Lerner Decl. ¶ 108 (“Differential network management and pricing arrangements have the potential to
generate significant benefits for consumers, competition, and innovation.”); id. ¶ 2 (“Imposing ex ante regulation
that restricts the ability of broadband providers to experiment and deploy such business models, without a clear
showing that such provider conduct harms the competitive process, would distort market outcomes, reduce
investment incentives, and thereby harm consumers. This is particularly true given the rapidly changing and
dynamic nature of the industry, and the significant uncertainty concerning the services and business models that will
best serve consumers going forward.”).
80
81
Katz Decl. ¶ 36.
Id. ¶ 111.
30
sided market arrangements or other sophisticated approaches to pricing.82 Some providers trying
to distinguish themselves in the competitive broadband marketplace already have begun
experimenting with such arrangements. T-Mobile’s Music Freedom plan, which exempts music
services from subscribers’ monthly data allowances, could benefit consumers who frequently
stream music. AT&T’s new sponsored data program likewise permits consumers to access
sponsored content without having that associated usage count against their monthly data
allowance.
Such flexibility to experiment with alternative arrangements not only can reduce costs to
end users while allowing them to access the content they demand, but also benefit edge providers
and spur continued investment in broadband infrastructure.83 Differentiated arrangements help
content and edge providers distinguish themselves in the marketplace, as T-Mobile is trying to
do, and tailor their services to consumer demands. Indeed, the flexibility created by
differentiated arrangements allows content and edge providers to adapt their services quickly in a
competitive marketplace. And from the perspective of broadband providers, such arrangements
could help fund the ongoing heavy costs of infrastructure deployment and upgrades without
increasing costs to consumers. Given the potential benefits all around, such arrangements should
be permissible and should be tested in light of consumer demand rather than resolved through
regulatory wrangling. As explained below, allowing such arrangements would not foreclose the
Commission’s ability to address particular harmful practices on an appropriate record.84
An approach allowing meaningful flexibility for arrangements in addition to customers’
selected Internet access service would also help the Commission’s rules to pass legal muster. In
82
See also Katz Decl. ¶¶ 54-61.
83
See Katz Decl. ¶ 56 (noting that rationale for differentiated arrangements “is competition on the merits”).
84
Infra Section II.D.
31
upholding the Commission’s data roaming rules, the D.C. Circuit explained in Cellco that the
Commission cannot “in practice” “relegate[] [broadband providers] to common carrier status.”85
The Court emphasized two keys to avoiding common-carrier restrictions. First, any rule must
leave “substantial room for individualized bargaining and discrimination in terms”86—a
commercial reasonableness rule cannot amount to a duty to “hold out facilities indifferently for
public use.”87 Second, the Commission should leave “considerable flexibility for providers to
respond to the competitive forces” of the market.88 The data roaming rules, the Court concluded,
“ensure[d] providers more freedom from agency intervention”89 than the pro tanto common
carriage restrictions invalidated in Midwest Video II, which “delimit[ed] what [providers could]
charge for access and use of equipment.”90 The Court concluded by observing that “the
Commission would . . . do well to ensure that the discretion carved out in the rule’s text remains
carved out in fact.”91 Prescriptive rules that reduce flexibility thus would not be appropriate.
In order to ensure general flexibility, therefore, any commercial reasonableness standard must, in
practice, “differ materially” from common carriage.92 As the D.C. Circuit made clear in Verizon,
even common carriers traditionally had the right to turn away business due to technological
constraints, economic or technical unfeasibility, and the like, and broadband providers must have
85
Cellco P’ship v. FCC, 700 F.3d 534, 547 (D.C. Cir. 2012) (alteration omitted).
86
Id. at 548.
87
FCC v. Midwest Video Corp., 440 U.S. 689, 706 n.16 (1979) (emphasis added) (“Midwest Video II”).
88
Cellco, 700 F.3d at 548; see Verizon v. FCC, 740 F.3d 623, 657 (2014) (invalidating the Open Internet
Order’s anti-discrimination rule partly for lack of flexibility).
89
Cellco, 700 F.3d at 548.
90
Id. at 547, quoting Midwest Video II, 440 U.S. at 702.
91
Cellco, 700 F.3d at 549.
92
Id. at 547.
32
at least as much flexibility to decline to enter into differentiated arrangements.93 If parties cannot
reach an agreement, for example, that should be permissible so long as the broadband provider
acts in a commercially reasonable manner and continues to provide the requisite tier of Internet
access service selected by the customer.
Using safe harbors and rebuttable presumptions would be another way to ensure a proper
distance between the Commission’s commercial reasonableness standard and common carriage
and provide greater certainty in this area. “Unless clear and meaningful safe harbors or
presumptions are identified, a rule prohibiting “commercially unreasonable” practices could
distort and discourage network management, innovative pricing, and other business practices that
would otherwise benefit consumers.”94
One such area where a safe harbor or presumption of reasonableness is appropriate is in
the case of non-exclusive arrangements with unaffiliated third parties. As a general matter, such
arrangements are unlikely to harm consumers or competition, and the Commission should
presume that such arm’s length arrangements are reasonable. As Professor Katz notes: “If a
deal is offered on a non-exclusive basis, then other edge providers can avail themselves of that
deal if they believe it will allow them to better differentiate their services or in some other way
compete more effectively. The non-exclusivity would ensure that the Internet access provider
could not ‘leverage’ any market power that it might possess to harm competition in the relevant
market for edge provider services.”95 Non-exclusive agreements with unaffiliated third-parties
help providers better adapt to consumer demand and only help, not harm competition.
93
See Verizon, 740 F.3d at 657 (discussing NARUC v. FCC, 525 F.2d 630, 641 (D.C. Cir. 1976) (“NARUC
I”)).
94
Katz Decl. ¶ 3.
95
Id. ¶ 51.
33
In addition, the Commission should address other specific areas where safe harbors
would be especially appropriate. Sponsored data and other arrangements that address only
pricing, for example, do not result in differential treatment of traffic over a customer’s
connection. But such arrangements could make service cheaper for end-users, enabling them to
access more content when and where they want it, and could provide a way for interested content
providers to promote or encourage use of their services.96 These and other arrangements that do
not entail the differential handling of traffic over the last mile – the main concern raised by
supporters of net neutrality regulation – should be presumed lawful.97 “The Commission should
not intentionally or unintentionally create barriers to such pricing by leaving it under a cloud of
regulatory uncertainty. Instead, the Commission should declare two-sided pricing to be
presumptively reasonable. Moreover, this presumption should be rebuttable only if the
challenging party can prove concrete harm to competition from the specific practice being
challenged.”98
The Commission’s presumptions likewise should take into account the important effect
of competition on preventing anti-consumer or anticompetitive conduct. As Professor Katz
notes: “In areas where consumers can choose among competing broadband Internet access
providers—particularly if more than one provider is offering very-high-speed access services—
there is less concern that an access provider will adopt policies that harm competition among
edge providers because attempts to do so may trigger consumers to switch access providers. In
addition to having less concern about the competitive effects of differentiated arrangements,
there is greater concern that regulation will impose costs by reducing flexibility and discouraging
96
See id. ¶¶ 54-61 (explaining consumer and competitive benefits of two-sided pricing).
97
See Lerner Decl. ¶ 110 n.157.
98
Katz Decl. ¶ 55.
34
experimentation.”99 Similarly, practices undertaken by providers in competitive marketplaces –
or practices previously found by the Commission to be commercially reasonable – should be
presumed reasonable.100
Finally, as explained below, if the Commission had a firm basis to conclude that
particular types of arrangements would harm consumers or competition, it could consider a
rebuttable presumption that such agreements are commercially unreasonable.101
In addition to such safe harbors and presumptions, any test for commercial
reasonableness should allow flexible consideration of whether a provider’s decision to enter a
particular differentiated arrangement (or not) causes any anti-competitive effects or harm to
consumers, just as in the data roaming context.102
Although Internet access is a different
service, certain factors relied on in the data roaming context also apply here and would help to
ensure that any new rule remains within the bounds of Cellco.103 By way of illustration, key
factors under a commercial reasonableness standard for fixed broadband might include:

the technical feasibility of entering into a differentiated service offering for the
content, application, service, or device for which it is requested;104

whether there are other options for securing a differentiated service offering from
other providers of broadband Internet access service;

the existence of widely adopted industry practices or applicable best practices or
standards adopted through multi-stakeholder groups;105 and
99
Id. ¶ 52.
100
Id. ¶ 53.
101
See infra Section II.D.
102
See Katz Decl. ¶¶ 37-40.
103
See Reexamination of Roaming Obligations of Commercial Mobile Radio Service Providers and Other
Providers of Mobile Data Services, Second Report and Order, 26 FCC Rcd 5411, ¶ 86 (2011) (“Data Roaming
Order”); Verizon, 740 F.3d at 657; Cellco, 700 F.3d at 548.
104
See Katz Decl. ¶ 39.
105
Id. ¶ 40.
35

the impact of the terms and conditions on the incentives for a provider of broadband
Internet access service to invest in facilities and coverage, services, and service
quality.
The ultimate touchstone, however, should be broadband providers’ flexibility “to respond to the
competitive forces” of the broadband Internet access service market, so long as their actions do
not harm competition or consumers.106 The Commission should reject regulations that would
inhibit innovation and investment and supplant the protections of the market in favor of static
and prescriptive rules. Technology, consumer needs, and business models are evolving at
lightning speed, and any effort to draft rules based on today’s circumstances inevitably would
backfire. A mandate dictating permissible forms of service differentiation, for instance, would
be outdated upon adoption because new technologies, service offerings, and demands would
quickly shift understandings of what was reasonable and how consumer needs would best be met
going forward.
D. Under the Proposed Commercial Reasonableness Rule, the Commission Could
Address Particular Practices That Harm Consumers or Competition.
Under the Commission’s proposed approach, the Commission could address particular
practices that harm competition or consumers so long as it also allows sufficient flexibility for
broadband providers to develop other differentiated arrangements. Indeed, the Commission can
address practices that harm consumers or competition under the appropriate record without
necessarily straying into impermissible common carriage or taking the radical step of
“reclassification.”
To take one example that has received much public attention, some parties have raised
concern about the potential harms of “paid prioritization” arrangements. If the Commission were
to determine that certain types of paid prioritization do, in fact, harm competition or consumers,
106
Cellco, 700 F.3d at 548.
36
it could prohibit such arrangements without straying into impermissible common carriage so
long as it left adequate room for other types of differentiated arrangements.
Just because the Commission could prohibit such practices in some contexts, does not
mean that it should do so at the present time and instead could wait to see whether any such
theorized harms actually materialize. Again to return to the example of paid prioritization,
notwithstanding the recent attention, it is far from clear that there is a problem to be addressed.
To place this in context, neither Verizon nor any other broadband providers of which we are
aware has introduced any form of paid prioritization arrangement to date, nor expressed a public
interest in doing so. Verizon has no plans for such a service, and it is unclear – particularly
given the widespread use of CDNs and other innovative technical means to ensure high-quality
transmission of content and the ever-improving capabilities of broadband networks – that there
would be much benefit to most Internet traffic from prioritization, particularly for the “big guys”
on the Internet. To improve service for their customers, large Internet providers and others have
already, in effect, created “fast lanes” for themselves through these alternative means such as
their own private networks and CDNs. The light-touch approach encouraged the development of
CDNs and similar ways to improve service for consumers and facilitate additional types of
Internet traffic. But in light of such existing approaches, many of the big Internet players are
unlikely to have much interest in paying for prioritization even if that were an available option
and the potential beneficiaries would more likely be smaller Internet players looking to compete
against such incumbents.
Likewise, there has not been a material level of demand for prioritization for typical
subscribers’ Internet usage thus far. That is because, while prioritization is potentially useful in
particular contexts – where, for instance, there is a need for especially low latency – it does not
37
provide much benefit for most Internet traffic. This becomes all the more true as the capabilities
of last-mile facilities improve and as content providers and others rely more heavily on
innovative arrangements such as content delivery networks, localized content storage and other
similar approaches that improve the speeds and performance with which content reaches
subscribers. In addition, best efforts broadband Internet access is becoming consistently faster
and cheaper.107 Under these circumstances, where the prospect and effects of a practice are
unclear, a wait-and-see approach is preferable.
To the extent that the Commission still has concerns about paid prioritization, however, it
could address particular practices without impermissibly imposing common carriage or taking
the dangerous step of “reclassifying” broadband, so long as it leaves room for other
individualized arrangements. On an appropriate record demonstrating that certain paid
prioritization practices have clear anti-competitive or anti-consumer effects, the Commission
even could create a rebuttable presumption that those specific practices are unreasonable –
without lapsing into common carriage. The key would be to retain other types of individualized
agreements beyond paid prioritization, such as sponsored data, two-sided pricing, or other benign
arrangements.
III.
ADDITIONAL REGULATION IS ESPECIALLY UNWARRANTED IN THE
MOBILE BROADBAND MARKETPLACE, WHICH PRESENTS UNIQUE
CIRCUMSTANCES AND IN WHICH AMERICA LEADS THE WORLD.
As the Commission has previously recognized, mobile broadband presents special
considerations warranting an especially flexible approach to open Internet rules.108
107
Supra n. 33.
108
See Open Internet Order ¶ 94.
38
A. Unique Features Distinguish Mobile Broadband Service from Fixed Broadband
Service.
Mobile wireless broadband services are provisioned in an especially innovative and
competitive environment – one that is still developing quickly – and present unique technical
challenges that distinguish them from fixed broadband offerings.
Mobile wireless broadband services are rapidly developing. Since 2010, the mobile
broadband market has continued to evolve at breakneck speed as providers invest billions of
dollars in their wireless networks, prompting developers of applications and devices to respond
with new offerings to take advantage of those enhanced network capabilities.109 The results
speak for themselves: By mid-2013, the share of Americans with access to wireless broadband
download speeds of greater than 10 Mbps had increased to over 95 percent,110 and increased
bandwidths have led to newer and better applications. Apple’s App Store and Google Play
currently both offer over one million apps for consumers to download for iOS and Android
mobile devices respectively.111 Since the release of the 2010 Open Internet Order, the wireless
industry has seen a staggering increase in data consumption, driven by the widespread
deployment of 4G service.112 Verizon has already matched its 3G footprint and offers 4G service
to more than 300 million Americans. We recently took steps to significantly increase capacity in
109
See Ericsson Mobility Report, at 10-12 (Nov. 2013), http://www.ericsson.com/res/docs/2013/ericssonmobility-report-november-2013.pdf (“Ericsson Mobility Report”); CTIA, The U.S. Wireless Industry: Leading the
World in Investment, Value, Innovation, and Competition, at 15-16 (Nov. 2013) (“Leading the World”), attached to
Letter from Scott K. Bergmann, CTIA, to Hon. Thomas E. Wheeler, et al., FCC, GN Docket No. 09-51, WT Docket
No. 13-135 (Nov. 13, 2013).
110
See Lerner Decl. ¶ 82, citing Federal Communications Commission National Broadband Map,
http://www.broadbandmap.gov/summarize/nationwide. .
111
See Lerner Decl. ¶ 80, citing Nathan Ingraham, Apple Announces 1 million apps in the App store, more
than 1 billion songs played on iTunes radio, THE VERGE (Oct. 22, 2013, 1:12 PM)
http://www.theverge.com/2013/10/22/4866302/apple-announces-1-million-apps-in-the-app-store; Josh Ownby,
Google Play Store Hits Million-Apps Milestone, THE HORN (Aug. 14, 2013),
http://www.gazelle.com/thehorn/2013/08/14/google-play-store-hits-million-apps-milestone.
112
See Leading the World at 2.
39
order to improve consumers’ experience, and are already working on the next generation of
mobile broadband service. Network investments by Verizon and other mobile broadband
providers have made the United States the world leader in wireless innovation, placing it at the
epicenter of the app economy and the global device marketplace.113 Exploding demand for
mobile data is likely to continue as Verizon and other mobile providers augment their LTE
networks and begin to look at next-generation capabilities such as LTE Advanced.114 Cisco, for
example, predicts that mobile data traffic in the United States and Canada will grow eightfold
between 2013 and 2018 – a rate three times faster than the growth of fixed IP traffic over that
same period.115 The growth of the “Internet of Things” is contributing to the rapidly growing
demand for mobile data, and will continue to do so, likely spurring additional network
enhancements: By 2020, the number of connected devices is expected to grow nearly 30-fold to
26 billion units globally, a figure more than three times the number of smartphones, tablets, and
PCs.116
Although wireless broadband services have grown and evolved tremendously, the mobile
Internet is still a work in progress that continues to develop and evolve in critical ways.117 For
example, mobile broadband services currently compete with fixed services to some extent, and
have achieved speeds more than sufficient for many uses of broadband. But, so long as
continued innovation is encouraged and permitted, mobile broadband offerings have the potential
113
American consumers can now choose from among more than 790 handsets and devices that feature over a
dozen different operating systems, as offered by multiple facilities-based service providers and resellers and more
than 50 different device manufacturers. Leading the World, at 15-17.
114
See Ericsson Mobility Report at 10-12; Leading the World at 15-16.
115
See Cisco Systems, VNI Mobile Forecast Highlights, 2013 - 2018 (see “Filter by Region”, select “North
America”), http://www.cisco.com/assets/sol/sp/vni/forecast_highlights_mobile/index.html.
116
See Gartner Says the Internet of Things Installed Base Will Grow to 26 Billion Units By 2020, (Dec. 12,
2013), http://www.gartner.com/newsroom/id/2636073.
117
See Katz Decl. ¶ 62-63.
40
to compete with fixed services even more directly over time as providers find new ways to boost
speeds and capacity.118 As Professor Katz notes: “[T]he value of investments in innovative
network technologies, such as 5G, that promise to deliver dramatically higher speeds to
individual users are particularly large. Avoiding inflexible rules that discourage this innovation
and investment will increase the degree to which mobile broadband Internet access services
provide an effective competitive alternative to fixed-line services.”119
The mobile wireless broadband marketplace is intensely competitive. The wireless
broadband market is extraordinarily competitive, offering consumers a great multiplicity of
choices.120 As the Commission found in the most recent Mobile Competition Report, 82 percent
of American consumers can choose from at least four wireless broadband providers, while
roughly 98 percent can choose from at least two.121 Wireless broadband providers’ conduct
reflects the competitive marketplace: As of December 2013, wireless providers had made over
$398 billion in cumulative capital expenditures to build and upgrade their networks, reflecting
9% year-over-year growth since 1985.122 Indeed, wireless providers have invested four times
118
See, e.g., Lerner Decl. ¶ 84 (“As wireless broadband providers continue to improve their networks to
increase speeds, and as more wireless spectrum becomes available, wireless networks are likely to become an
alternative to wireline providers for a larger range of broadband users and uses.”); id. ¶ 86 (“Although the wireless
broadband industry is still nascent, and evolving rapidly, these industry developments suggest that wireless networks
exert some degree of competitive pressure on wireline broadband providers. With continued, rapid innovation and
investment, there is the opportunity for wireless broadband to be a more attractive competitive alternative over
time.”).
119
Katz Decl. ¶ 3.
120
See id. ¶ 62.
Mobile Competition Report, at Table 9.
121
122
See CTIA-The Wireless Association, Annualized Wireless Industry Survey Results - December 1985 to
December 2013, (June 17, 2014), http://www.ctia.org/docs/default-source/FactsStats/ctia_survey_ye_2013_graphics-final.pdf?sfvrsn=2 (last visited July 10, 2014). See also Lerner Decl. ¶ 72.
41
more in their networks per subscriber than the global average.123 Meanwhile, wireless data
prices have plummeted.124 These developments are the hallmarks of a highly competitive
marketplace.
Wireless broadband services face unique technical and operational constraints.
Wireless broadband services face technological and operational constraints arising from the need
to manage spectrum sharing by a dynamically varying number of mobile users at any time. As
the Commission has recognized, “conditions in mobile broadband networks may necessitate
network management practices that would not be necessary in most fixed networks.”125 For
example, wireless broadband communications require complex and dynamic management of
spectrum as the number and mix of users being served by a cell site changes in sometimes highly
unpredictable ways. Moreover, these complexities are compounded by the scarce spectrum
resources available to each provider, which constrain a provider’s ability to increase network
capacity to meet subscriber needs and quality-of-service expectations. Wireless networks face
limits imposed by finite spectrum resources. While wireless providers can and do innovate to
maximize the traffic that can be transmitted over the available spectrum, they cannot make more
of it, and must instead manage their networks to overcome spectrum scarcity.
123
See Didier Scemama, et al., 2014 wireless capex: BRICs & Europe to pick up the slack, Bank of America
Merrill Lynch, Global Telecom Equipment, Jan. 13, 2014, at Table 2. See also Glen Campbell, 2014: The year
ahead, Bank of America Merrill Lynch, Global Wireless Matrix 4Q13, Jan. 8, 2014 at Tables 1 & 2. See also
Lerner Decl. ¶¶ 69-72.
124
See Maeghan Ouimet, INFOGRAPHIC: The Staggeringly Huge Future of Mobility, VISAGEMOBILE
MOBILITY BLOG, (Sept. 6, 2012), http://visagemobile.com/mobilityblog/2012/09/06/infographic-the-staggeringlyhuge-future-ofmobility/.
125
Open Internet Order ¶ 103.
42
B. Mobile Wireless Broadband’s Unique Features Warrant a Particularly LightTouch Regulatory Approach.
Given the dynamic innovation, intense competition, and unique technical constraints that
characterize the mobile broadband marketplace, prescriptive regulation would do particular harm
to mobile broadband customers. Mobile broadband remains a nascent technology, and will only
become a true competitor to fixed service following substantial investment, development, and
innovation. “Accordingly, providing flexibility for wireless broadband providers to experiment
with and implement different service offerings and business models is particularly important to
enhancing the competitiveness of both the wireline and wireless industries.”126 However wellintentioned they might be, prescriptive rules would limit the technical flexibility that is so critical
to the high-quality mobile services that subscribers demand. Moreover, rigid requirements
would stymie rapid innovation in the mobile broadband ecosystem even more than in the context
of fixed service, further constraining operators’ efforts to diversify their offerings and address
consumers’ evolving preferences. Such rules could even alter the trajectory of mobile wireless’s
development, precluding the emergence of even stronger competitors to fixed broadband
offerings.127 Additionally, in the mobile context, decisions made by the developers of operating
systems, devices, applications, as well as app store operators, often dictate whether a particular
service or application is available and will work for a particular customer. Applying wireline
rules to mobile broadband providers could confuse consumers and create false expectations of
126
Lerner Decl. ¶ 15; see also Katz Decl. ¶¶ 3; 62-63.
127
See id. ¶ 3.
43
compatibility. There is no reason to risk these adverse consequences when the competitive
market has protected consumers so well, and will continue to do so.128
If the Commission nevertheless opts to apply new requirements for mobile broadband, it
should retain the more flexible approach reflected in the 2010 Open Internet Order and the
NPRM. For example, with respect to blocking, the Commission should adopt the NPRM’s
tentative conclusion to maintain the 2010 framework, which prohibited mobile broadband
providers from blocking access to lawful websites or competing voice and video telephony
applications.129 The Open Internet Order appropriately recognized that the download and use of
a mobile application presents unique network management issues.130 For example, mobile
providers may use sophisticated queuing and scheduling algorithms that send more packets to
users during times when the “signal-to-noise” ratio is high and fewer packets when that ratio is
low.131 Factors of this type led the Commission to adopt a tailored no-blocking rule for mobile
wireless in 2010, and continue to warrant a targeted approach today.132
Likewise, consistent with the NPRM’s tentative conclusion, the Commission should
decline to apply additional rules, such as the “commercial reasonableness” requirement proposed
for fixed providers, to mobile broadband providers.133 These providers can be expected to
behave in commercially reasonable ways for the same reasons that companies in hundreds of
128
Mobile broadband is also helping fulfill other national goals. As the Commission found in 2010, mobile
broadband “is an important Internet access platform that is helping drive adoption.” Open Internet Order ¶ 94.
Mobile’s role in promoting adoption presents one more rationale for a light touch here.
129
See id. ¶ 105; NPRM ¶¶ 62, 94.
130
See Open Internet Order ¶ 100 (“We also recognize that accessing a website typically does not present the
same network management issues that downloading and running an app on a device may present.”).
131
See Comments of Verizon and Verizon Wireless, Framework for Broadband Internet Service, GN Docket
No. 10-127, at Attachment B, Declaration of Jeannie H. Diefenderfer and Thomas K. Sawanobori, ¶ 20 (July 15,
2010) (“Verizon Broadband Services Comments”).
132
The Commission should also expressly reaffirm that network operators have the ability to offer parental
controls and other usage controls or to place a block on any device for any application that its consumers request.
133
See NPRM ¶¶ 62, 140. See also Katz Decl. ¶ 62.
44
other ultra-competitive markets do: because their customers will go elsewhere if they do not,
and because existing laws already bar unfair and deceptive business practices. Indeed, the
existing transparency rule renders mobile broadband customers even better able to respond to
unreasonable practices than consumers in most markets.
The question, then, is not whether mobile broadband providers have strong incentives to
behave reasonably – they do – but whether the Commission will subject their practices to a posthoc review process that does not apply even in far less competitive markets. It should reject this
latter course, and maintain the current, more flexible approach that continues to be critical to this
healthy and dynamic sector. As explained in the 2010 Open Internet Order134 and above, mobile
broadband services continue to evolve rapidly as providers develop innovative offerings to meet
consumers’ changing needs. Further restricting the practices of mobile broadband providers
would inhibit innovation and experimentation and decrease the number of options available to
consumers, causing the mobile market to become homogenized and reducing competition as a
result. It would also increase uncertainty in an environment where robust network management
practices are essential to meeting customers’ needs. A provider seeking to introduce an
innovative new offering – for example, a mobile medical monitoring service or a mobileoptimized content streaming offering – would be forced to weigh the risk that the offering might
later be deemed “unreasonable,” notwithstanding its popularity among consumers. Consumers
should not be limited by these risks, or confined to a single, government-sanctioned mobile
broadband experience. Rather, the Commission should leave mobile users free to choose from
among a broad range of different experiences, according to their preferences. Excluding mobile
134
See Open Internet Order ¶ 94.
45
broadband from the commercial reasonableness standard will ensure greater choice in the mobile
marketplace, maximizing consumer welfare.
IV.
BROADBAND RECLASSIFICATION WOULD BE A RADICAL AND RISKY
STEP THAT WOULD JEOPARDIZE NATIONAL GOALS AND VIOLATE THE
COMMUNICATIONS ACT.
As an alternative to using the Section 706 authority upheld by the D.C. Circuit to regulate
broadband Internet access service, the Commission asks whether it should reclassify that service
as a “telecommunications service,” in whole or in part.135 This approach would be a radical and
risky reversal of a successful policy that administrations have uniformly pursued on a bi-partisan
basis since the Clinton administration in the 1990s. The prospect of 19th-Century price
regulation and Title II’s other arcane requirements would stifle investment in and development
of the Internet. Yet Title II would not even address what reclassification supporters say are their
concerns—in other words, this unlawful approach is just a Trojan horse to obtain regulation for
regulation’s sake. And to the extent that proponents of reclassification want Title II to
straitjacket innovation, the only beneficiaries of reclassification would be big, established
Internet players: any resulting lack of flexibility would forestall the development of
differentiated arrangements that might help smaller players be more competitive in the face of
the significant advantages established Internet companies have created for themselves.
Reclassification not only would be a radical reversal of approach; it also would be
harmful, casting a regulatory pall over virtually the entire Internet ecosystem. The same theory
invoked to justify Title II here would apply equally to myriad other services that include or use
telecommunications. Reclassification also would have international ramifications that might
lead some countries to impose terminating access charges or rate regulations on U.S. broadband
135
NPRM ¶ 148.
46
providers and Internet companies. And as a purely legal matter, if anything, reclassification
would face a higher legal hurdle than the Commission’s prior rules and, at a minimum, would
result in years of investment-deterring litigation and uncertainty.
Even if the Commission sought to forbear from certain Title II regulations, that would
only exacerbate the complications and uncertainty associated with reclassification, as the
Commission and interested parties parsed through which of the ill-fitting provisions to keep and
which to toss. Indeed, some of the most restrictive provisions of Title II – including Sections
201, 202 and 214 on which most price and service regulation are founded – are those embraced
by the proponents of reclassification as essential to their cause; unless the Commission forbore
from these provisions, the bulk of the Title II problems would remain. The Commission should
not reverse decades of successful policy to risk what would be the most radical change to
Internet regulation since the creation of the Internet.
A. Reclassification Would Be a Radical and Risky Reversal of Successful Policy
Uniformly Championed by Commissions for Two Decades.
Imposing a Title II common carriage regime on broadband providers would be a radical
change in course that would only chill, not spur innovation. Title II is a regulatory dinosaur,
crafted eighty years ago – and based on 19th-Century laws regulating railroads – to address the
one-wire world of rotary telephones. All of the hallmarks of Title II – rate regulation,
mandatory fees, and the need for advance regulatory permission before offering or discontinuing
services – were tailored to address an environment characterized by a government-sponsored
monopoly for the provision of pure, relatively simple, and standardized transmission services
(i.e., rotary telephone service). That government-granted monopoly and the rudimentary service
it proffered would be a radical and risky new approach for today’s fast-paced and competitive
marketplace for broadband and the wide range of sophisticated services that it encompasses.
47
Every administration since the Clinton administration has recognized that Title II would
be dangerous in the dynamic broadband marketplace. Shortly after the 1996
Telecommunications Act, the Commission stressed, “In no respect are we considering regulating
the Internet. Rather. . .we seek to reduce barriers to competition so that companies in all
segments of the communications industry have the incentive to innovate.”136 In 2002, the
Commission formally classified broadband service as an information service, finding that
“broadband services should exist in a minimal regulatory environment that promotes investment
and innovation in a competitive market.”137 And again in its 2005 Wireline Broadband Order,
the Commission established “a minimal regulatory environment for wireline broadband Internet
access services to benefit American consumers and promote innovative and efficient
communications.”138
For nearly twenty years, the light-touch approach uniformly taken by successive
administrations has been successful in ensuring investment, experimentation, and explosive
growth in broadband capabilities and services.139 For example, that successful regulatory
framework has spurred Verizon and others throughout the Internet ecosystem to invest billions of
dollars in building out broadband networks and developing the services that ride on them. For
136
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, Report, 14 FCC Rcd 2398, ¶ 18 (1999).
137
Inquiry Concerning High-Speed Access to the Internet Over Cable and Other Facilities; Internet Over
Cable Declaratory Ruling; Appropriate Regulatory Treatment for Broadband Access to the Internet Over Cable
Facilities, Declaratory Ruling and Notice of Proposed Rulemaking, 17 FCC Rcd 4798, ¶ 5 (2002) (“Cable
Broadband Order”) (internal citation omitted).
138
Appropriate Framework for Broadband Access to the Internet over Wireline Facilities; et al., Report and
Order and Notice of Proposed Rulemaking, 20 FCC Rcd 14,853, ¶ 1 (2005) (“Wireline Broadband Order”).
139
See, e.g., id., ¶ 44; Cable Broadband Order, ¶¶ 44, 52-55; United Power Line Council’s Petition for
Declaratory Ruling Regarding the Classification of Broadband over Power Line Internet Access Service as an
Information Service, Memorandum Opinion and Order, 21 FCC Rcd 13,281 (2006) (“BPL Classification Order”);
Petition for Forbearance of the Verizon Telephone Companies Pursuant to 47 U.S.C. § 160(c); et al., Memorandum
Opinion and Order, 19 FCC Rcd 21,496, ¶ 19 (2004).
48
example, “[c]able operators responded to the deployment of fiber networks by also offering very
high-speed broadband services, including by rolling out DOCSIS 3.0 technology,” and quickly
extended their DOCSIS 3.0 deployment to reach about 85 percent of the population.140
Broadband providers have continued to improve the quality and capabilities of their Internet
access services, often integrating more advanced features to improve the security, performance,
and functionalities of their services. As a result, these services bear little resemblance to the pure
telephone transmission service for which Title II was created. Moreover, broadband providers
are now offering a wide range of cloud-based and over-the-top services, both for customers of
their Internet access services and increasingly to reach customers over other ISPs’ networks.
These sophisticated services all increase consumer choice.
Title II, by contrast, would cripple that freedom, flexibility, and innovation, for its core
provisions – such as intrusive price regulation and entry and exit regulation – are classic
examples of the kind of arcane regulations that deter investment. Price regulation under Section
201 would empower the Commission, not the market, to determine the value of broadband
Internet access. As the Department of Justice warned as recently as 2010, such price regulation
would threaten investment in broadband infrastructure and could “stifl[e] the infrastructure
investments needed to expand broadband access.”141
Title II’s requirement that providers obtain regulatory permission before introducing or
discontinuing services also stands in stark contrast to the “permission-less” innovation that has
been the hallmark of the Internet. If applied to Internet services, Section 214 would prohibit
broadband providers from adding or extending new services, or discontinuing old services,
140
Lerner Decl. ¶ 9.
141
Comments of the Department of Justice, Economic Issues in Broadband Competition; A National
Broadband Plan for Our Future, GN Docket No. 09-51, at 28 (Jan. 4, 2010).
49
without Commission pre-approval.142 Yet fast-paced innovation has been essential to market
success in the Internet ecosystem. If a provider develops a better way to serve its customers, it
races to roll out new features or updates quickly in order to stay ahead of the competition; and if
a new service fails to attract customers, a provider must be able to quickly withdraw the service
and move on to the next idea. If offering and withdrawing a service were to require FCC
approval, the moment for innovation likely would pass long before the Commission could act.
Applying Title II to the American broadband industry would be like tying a cinder block
to the ankle of an Olympic sprinter in the midst of a race and then wishing her luck.
Reclassification would create a major drag on new and improved broadband infrastructure, even
though substantial investment in such infrastructure is precisely what is needed to keep pace with
exponentially increasing consumer demands for bandwidth. By chilling such investment and
discouraging innovation, Title II and related proposals would only impede, not advance, the
public’s access to and enjoyment of the Internet. Broadband services and features would ossify,
become less robust, and be less able to meet consumers’ demands over time. It is no wonder that
previous administrations uniformly have avoided that radical path.
Because Title II is such a risky fit for broadband, its requirements would engender legal
confusion that could further chill innovation. Many Title II provisions cannot rationally be
applied to broadband providers at all. Section 223, for example, deals with obligations related to
obscene or harassing telephone calls; Section 226 with telephone operator services; Section 227
with restrictions on the use of telephone equipment.143 These and other Title II requirements
have no bearing whatsoever on Internet services, much less Internet openness. Reclassification
142
See, e.g., 47 U.S.C. § 214(a) (prohibiting “the construction of a new line or of an extension of any line,” or
“transmission over or by means of such additional or extended line, . . . unless and until there shall first have been
obtained from the Commission a certificate that the . . . public convenience and necessity require” such action).
143
47 U.S.C. §§ 223, 226, 227.
50
could therefore generate confusion and legal wrangling as the Commission and industry sorted
out how to address such requirements in this context.
Forbearance would not solve these problems. Even if the Commission sought to forbear
from certain provisions of Title II, that approach would cause an additional legal battle over the
scope of such action: which provisions do, or do not, apply, and why that is so. While
proponents of Title II suggest that forbearance is the answer to concerns about overly restrictive
regulation, the same statutory provisions that they embrace as essential to regulation of the
Internet – such as Sections 201, 202, and 214 – are the very provisions on which volumes of
price and service regulation are based. Unless the Commission was willing to forbear from these
pillars of Title II or to meaningfully limit their scope, the regulatory drag and uncertainty
imposed by Title II would persist. Thus, battles over the scope and reach of forbearance would
simply pile on to the complexity and uncertainty associated with reclassification, without
meaningfully reducing the risk of harmful over-regulation.
In sum, previous administrations were correct in uniformly recognizing that Title II is a
risky fit for the dynamic and competitive broadband Internet marketplace. Reclassification
would all but guarantee that exciting new Internet access services and powerful new
infrastructure become less, rather than more, available to consumers. It would be a radical move
that effectively would reverse the success the Internet has seen in the past two decades.
B. Reclassification Would Not Even Achieve the Results Most Desired By Its
Proponents – It Would Be Regulation for Regulation’s Sake.
While reclassification would pose considerable risks of disrupting the sustained
momentum that has made the Internet more robust and useful to consumers, it would not even
achieve the main results desired by its proponents: namely, banning two-sided arrangements or
any arrangements that would result in some bits being handled differently than others. By the
51
terms of the Communications Act, a “telecommunications service” is an offering of
“telecommunications for a fee directly to the public.”144 Thus, by reclassifying the service that
broadband providers afford edge providers as a separate “telecommunications service,” the
Commission would require that broadband providers charge edge providers for that service.145 In
fact, two-sided arrangements have existed for decades under Title II, as both customers and long
distance providers have contributed to the cost of providing local telephone service. Far from
preventing two-sided arrangements, reclassification would mandate them.
Moreover, the history of common carriage is replete with examples of the kinds of
differentiated services and pricing that reclassification proponents claim they want to prevent.
Title II does not prohibit all differentiation – it merely prohibits unreasonable discrimination in
“terms and conditions” for “like” services among “similarly situated” customers.146 The D.C.
Circuit has made clear that two services are not “like” if they are “different in any material
functional respect,”147 and differentiation among “like” services is permissible so long as it is
reasonable. Under this standard, the Commission has upheld a wide range of differentiated terms
and conditions such as:
Differentiated service levels – Title II authorizes common carriers to offer service level
agreements providing different qualities of service and different service level
guarantees, for their transmissions.148 Because the Commission has never prohibited
144
47 U.S.C. § 153(46) (emphasis added).
145
See, e.g., Qwest Comms. Co. v. Northern Valley Comms. Co., Memorandum Opinion and Order, 26 FCC
Rcd 8332, ¶¶ 7-10 (2011) (invalidating access service tariff under Section 201(b) where CLEC did not charge endusers for service).
146
Title II prohibits “unjust or unreasonable discrimination in charges, practices . . . or services,” but expressly
permits the creation of “different charges . . . for the different classes of communications” that “the Commission
may decide to be just and reasonable.” 47 U.S.C. § 201(b).
147
MCI Telecomms. Corp. v. FCC, 917 F.2d 30, 39 (D.C. Cir. 1990) (internal citation omitted).
148
See, e.g., Competitive Telecomms. Ass’n v. FCC, 998 F.2d 1058, 1064 (D.C. Cir. 1993) (“By its nature,
§ 202(a) is not concerned with the price differentials between qualitatively different services or service packages. In
other words, as far as ‘unreasonable discrimination’ is concerned, an apple does not have to be priced the same as an
orange.”).
52
carriers from offering customers the option of purchasing a higher quality of service,
that precedent under Title II would permit tiered pricing for “best efforts” and
premium delivery services;
Differentiated prices – Title II has long authorized common carriers to offer customerspecific “contract tariffs,” individualized pricing based on in-store customer
bargaining, and even “individual case basis” arrangements, such as 8XX service, that
shift the payment obligation from one party to another.149 Thus, Title II could permit
analogous arrangements, such as ad-based Internet access or individualized contracts;
Volume and term pricing – Title II permits common carriers to offer volume discounts
for customers who purchase services in larger volumes, and term discounts for longterm contracts. These sorts of arrangements likely would enable broadband providers
to offer tiered levels of data consumption, as well as discounts for long-term
contracts;
Prioritized installation and repair services – Title II authorizes common carriers to
prioritize installation and repair. This is yet another example of the ways in which
even Title II common carriers can offer differentiated services.
Reclassification therefore would not achieve the results its proponents want, but instead would
amount to regulation for regulation’s sake.
C. Any Resulting Lack of Flexibility Would Benefit Established, Not Small Players
in the Internet Ecosystem.
To the extent proponents of Title II regulation succeed in constricting the flexibility of
broadband providers, reclassification also would disserve proponents’ stated goal of helping the
proverbial innovator in her garage to compete against big, established Internet-based companies.
Bigger players already have created, in effect, their own fast lanes that smaller players cannot
readily replicate, including by constructing or contracting for extensive private networks, server
farms, content delivery networks and the like that improve the speed and reliability of their
content. To name just a few examples of such methods not realistically available to smaller
competitors: Google and other large Internet incumbents have deployed extensive private
149
See, e.g., Sea-Land Serv., Inc. v. ICC, 738 F.2d 1311, 1317 (D.C. Cir. 1984) (permitting carriers to
negotiate individual deals with customers at prices below tariff, so long as the same deal is made available to any
other customer seeking the same service at the same price).
53
networks and server farms around the globe to situate their content, such as YouTube videos,
closer to end-users and to give themselves significant performance advantages over competitors;
Netflix has built its “Open Connect” content delivery network to support its video service, and
until recently it denied the highest quality video to end users whose broadband providers did not
agree to host Netflix’s servers directly on their networks; and Google directs more traffic to its
proxy servers and network in connection with its particular implementation of the SPDY
protocol. This is not to suggest that such actions by established Internet players are inherently
suspect: to the contrary, many are innovative and may benefit consumers. But to ignore these
significant advantages is to ignore the reality of today’s Internet.
Broadband providers’ flexibility to offer differentiated options to other Internet players,
alongside the traditional “best efforts” Internet services that will continue to provide an
unrestricted path to consumers, could prove crucial to help smaller companies who may seek to
compete with the “fast lanes” and other advantages already enjoyed by the likes of Google,
Amazon, and Netflix over the rest of the Internet. With flexibility to develop differentiated
arrangements, broadband providers would have the ability to work with smaller providers to
develop service solutions that potentially could be beneficial and might help them to compete
with the established Internet giants in much the way that application stores have helped small
software developers compete with established software firms. Flexibility and innovation are
therefore essential to help level the playing field for less-established players in the Internet
ecosystem.
Ossification under Title II, therefore, would benefit the established Internet players in
comparison to smaller start-ups that would compete with them. Title II would eliminate
broadband providers’ flexibility to offer truly individualized arrangements and interconnection,
54
and it would severely slow their ability to develop other technological solutions quickly enough
to enable smaller companies to keep pace with the fast lanes of big players.
D. Reclassification Is a Dangerous Approach that Could Sweep in a Broad Range of
Internet Services.
Reclassification would be not just risky, but outright harmful, for the stultifying effect of
Title II would not necessarily be limited to the operators of broadband networks.
Reclassification could sweep in a broad range of Internet services, such as content, application,
and search providers, as well as e-Readers, GPS systems, and other devices. The impact of this
potential extension of Title II to additional quarters of the Internet would fundamentally change
the Internet as we know it, and not for the better.
The Commission long ago recognized that, if common-carriage regulation applies to
Internet access service, “it would be difficult to devise a sustainable rationale under which all, or
essentially all, information services did not fall into the telecommunications service category.”150
Like broadband, “essentially all” Internet-based services involve some combination of data
processing and data transmission – that is, both information service and telecommunications
service inputs. The Supreme Court has therefore warned that expanding the definition of
“telecommunications service” to include broadband would sweep in all Internet-based services
that “use telecommunications as an input to provide information service to the public.”151
Thus, if Title II were extended to Internet access services, it is unlikely that such a
decision ultimately would be confined to only a narrow sector of the Internet ecosystem.
Content, application, and search providers – such as Google and Netflix – use their own network
facilities or those of their vendors to speed the transmission of content of a user’s choosing to
150
Federal-State Joint Board on Universal Service, Report to Congress, 13 FCC Rcd 11,501, ¶ 57 (1998)
(“Stevens Report”).
151
See NCTA v. Brand X Internet Servs., 545 U.S. 967, 994 (2005).
55
and from that user. Google’s use of proxy servers in relation to its use of the SPDY protocol
likewise routes traffic over its network and through its routers both to speed up the loading of
web pages and especially to drive more traffic onto its routers and capture more customers’
information. And e-Readers and GPS devices include transmission capabilities to send and
upload content onto the device. There is no principled distinction between these information
services and broadband Internet access service.
Indeed, all manner of players in the Internet hold out services that include the
transmission of data: VoIP providers, online advertising providers, e-commerce providers, and a
wide range of other “over-the-top” providers, among others, would be at risk of similarly
burdensome regulation. As with broadband providers, these entities have relied on the freedom
to experiment with new technologies and business models. Subjecting them to Title II’s
lumbering approach of central government oversight and pre-approval would handicap the
development of Internet-related services – ironically, hampering the creation of new content and
applications that Title II proponents seek to promote.
Reversing course in favor of Internet regulation also would have significant international
ramifications. It would set a dangerous precedent at a time when the United States has needed to
fight vigilantly against international bodies and even repressive regimes that seek greater control
over the Internet. In countries where over-regulation of Internet access has already stunted
investment in broadband infrastructure, a radical shift in policy here could encourage foreign
leaders to extend that regulation to Internet companies of all types. For example, some countries
are now targeting large Internet companies, many of which are based in the United States, and
are looking to impose government-mandated terminating access charges on data traffic from
those U.S. providers. Reclassification would therefore send the wrong signal at the wrong time.
56
E. Reclassification of Broadband Service, in Whole or in Part, Would Be Unlawful,
and the Legal Uncertainty Would Chill Innovation.
In addition to its disastrous results, the Title II approach would also be unlawful. If
anything, reclassification would face a much higher legal hurdle than the Commission’s prior
rules, and would, at a minimum, result in years of investment-deterring litigation.
Reclassification would require revisiting the basic fact that broadband Internet access service is
an information service, not an offering of the pure transmission of data. Yet now, more than
ever, broadband service is a tightly integrated service that cannot realistically be separated into
distinct telecommunications and processing components. Reclassification would therefore do
serious violence to the basic statutory distinction between “telecommunications service” and
“information service,” that underpins the 1996 Telecommunications Act. Furthermore, because
the Commission cannot compel common carrier status as either a statutory or constitutional
matter, reclassification in whole or in part will engender legal uncertainty that could chill
innovation for years to come.
Reclassification Would Face a Heightened Legal Standard. The Communications Act
defines “telecommunications service” as an offering of the transmission of data for a fee
“without change in the form or content of the information” – in other words, a pure transmission
service.152 Any service that changes, stores, or “process[es]” data is, by contrast, an “information
service.”153 Congress intended, as the Commission has repeatedly found, these categories to be
mutually exclusive.154 Accordingly, a “hybrid” or “mixed” service – which is classified
152
47 U.S.C. § 153(43) (defining “telecommunications”) (emphasis added).
153
Id. § 153(20) (defining “information service” as a service that offers the “capability for generating,
acquiring, storing, transforming, processing, retrieving, utilizing, or making available information via
telecommunications”).
154
See, e.g., Cable Broadband Order ¶ 41 (“the Act’s ‘information service’ and ‘telecommunications service’
definitions establish mutually exclusive categories of service”); Stevens Report, ¶¶ 13, 39, 43 (describing Congress’
intent and legislative history indicating that definitions are mutually exclusive).
57
according to consumer perceptions of the finished product155 – is by “regulatory necessity” an
information service whenever its data transmission components are integrated with data
processing components.156 As the Supreme Court held in Brand X, broadband service is properly
classified as an information service precisely because it transmits data only in connection with
further processing, and because that transmission is necessary to providing Internet access
service.157
To change course now and determine, as a factual matter, that broadband service is
actually not a single integrated offering would require heightened justification under FCC v. Fox
Television Stations, Inc.158 In Fox, the Supreme Court held that, where a “new policy rests upon
factual findings that contradict those which underlay [the agency’s] prior policy; or when its
prior policy has engendered serious reliance interests that must be taken into account,” an agency
must “provide a more detailed justification” than if it were acting on a blank slate.159 “It would
be arbitrary or capricious to ignore such matters,” the Court explained, because “a reasoned
explanation is needed for disregarding facts and circumstances that underlay or were engendered
by the prior policy.”160 Reclassification would implicate both scenarios by overturning numerous
155
See, e.g., BPL Classification Order, ¶ 38 (finding that whether a service also includes a
telecommunications “offering” “turns on the nature of the functions the end user is offered”); id. ¶ 35.
156
Brief for the FCC at 26-27, NCTA v. Brand X Internet Servs., 545 U.S. 967 (2005) (04-277, 04-281), 2005
U.S. S. Ct. Briefs LEXIS, at **45-46 (“FCC Brand X Br.”); see Stevens Report ¶ 13 (finding such an interpretation
necessary to ensure that “information service providers are not subject to regulation as common carriers merely
because they provide their services ‘via telecommunications.’”).
157
See Brand X, 545 U.S. at 998.
158
556 U.S. 502 (2009).
159
Id., 556 U.S. at 515 (2009) (majority opinion); see id. at 549 (Breyer, J., dissenting) (requiring “a more
complete explanation than would prove satisfactory were change itself not at issue” when the agency rested its prior
position either on “particular factual findings” or “its view of the governing law”).
160
Id. at 515-16 (majority opinion).
58
Commission findings that broadband’s integrated offering is an information service,161 and by
uprooting the reliance interests of players throughout the Internet ecosystem who invested
billions in building broadband networks and developed services premised on light-touch
regulation.162
The Commission cannot satisfy Fox’s heightened standard because, if anything,
broadband providers include more integrated and complex functionalities to improve the
security, reliability, and performance of their services today than they did in 2005 when Brand X
was decided. Over the ensuing decade, broadband services have continued to evolve and
become increasingly sophisticated, due in large part to the Commission’s policy of light-touch
regulation, as explained above. Thus, broadband Internet access services continue to integrate
information services, including data storage or email services that involve storing or utilizing
data; parental controls and other security functions that store security preferences, then filter data
as it is retrieved or generated by the consumer; and services for personalizing home portal pages
through generating or transforming information.163 In addition, broadband providers such as
Verizon now integrate into their broadband offerings ever-more advanced features and
capabilities, including cloud-based services for storing information, as well as for retrieving and
acquiring information via software services;164 new spam filters and other reputation systems for
161
See Appropriate Regulatory Treatment for Broadband Access to the Internet Over Wireless Networks,
Declaratory Ruling, Declaratory Ruling, 22 FCC Rcd 5901, ¶ 1 (2007); BPL Classification Order, ¶ 1; Wireline
Broadband Order ¶ 12; Cable Broadband Order ¶ 41.
162
See supra 10-16.
163
See Verizon Broadband Services Comments, at Attachment A, Declaration of Michael A. Ritter, ¶¶ 6-8.
164
See, e.g., Verizon Introduces Per-Hour Billing for Oracle Database and Oracle Fusion Middleware on
Verizon Cloud (Jan. 10, 2014), http://newscenter.verizon.com/corporate/news-articles/2014/01-10-per-hour-billingfor-oracle/.
59
processing potentially harmful data;165 and caching servers and CDNs that store media content to
enable consumers to access that content at faster speeds. From the perspective of a consumer,
then, information service capabilities and data transmission are more intertwined in broadband
Internet access service than ever before.
In fact, the services offered today tightly integrate data transmission with data processing
– so much so that the two cannot realistically be separated at all. Consider, for example, how an
end user even accesses a website by clicking on a hyperlink or typing in a domain name. Neither
the HTML tag used to encode the hyperlink nor the end user typing in a URL address identify
the target website by IP address – in this respect, the end user normally does not even know
where it wants to transmit information. Instead, the broadband provider transmits the URL of
the target website to its own servers, which house a DNS directory of domain names cross-listed
with IP addresses; the provider then searches that directory and generates the appropriate IP
address to query. As both the Commission and the Supreme Court have recognized, the
transmission of data here cannot be separated from the processing that occurs with the DNS
service because, without that processing, the provider would not even know where to transmit
the information.166 Such processing occurs every time an end user clicks on a link or types in a
domain name.
Now consider what happens when the edge provider tries to send content to the end user.
Although the great majority of Internet IP addresses still use the IP address system called IPv4,
165
See Verizon Broadband Services Comments, at Ritter Decl., ¶¶ 12-17, 24-25; id. at
Diefenderfer/Sawanobori Decl.,¶ 8.
166
See Cable Broadband Order ¶ 38; Brand X, 545 U.S. at 998-1000.
60
increasingly many at-home consumer devices have addresses on the new IPv6 address system.167
If an edge provider uses a router programmed for IPv4 addresses, that router will be unable to
direct a packet to an end user with an IPv6 address. To correct this problem, broadband
providers have begun to maintain servers with dual IPv4 and IPv6 addresses that transform an
IPv4 address into an IPv6 address, and vice-versa. Without this necessary processing
component, data from an edge provider on the IPv4 Internet could not be transmitted to an end
user on the IPv6 Internet.
Countless other features of broadband Internet access service inextricably combine data
transmission and data processing. Security features, for instance, operate by having a broadband
provider “tag” certain IP addresses as security risks and redirecting a user to a safe filter page
whenever the user tries to access those sites. Similarly, parental controls create filters that block
certain content on a user’s device by having the broadband provider identify, mark, and
ultimately block unwanted content as it travels over the network. And cloud-based services
allow remote data storage and access to computing software where the actual computing is done
via communication between the end user’s computer and a data center server housed by the
broadband provider. In all of these features of broadband service, no data transmission can occur
without concomitant data processing: the two are inseparable.
Reclassification Would Violate the Communications Act. Because the
telecommunications component of broadband service is inseparable from its data processing
component, reclassification would contravene Congress’ express determination that a carrier be
167
Comcast was the first broadband provider to begin offering IPv6 addresses to at-home consumer devices, in
2011. See Iljitsch van Beijnum, Comcast begins limited IPv6 rollout to home customers, ARSTECHNICA (Nov. 10,
2011 4:51 PM), http://arstechnica.com/business/2011/11/comcast-begins-limited-ipv6-rollout-to-home-customers.
In 2010 less than 1% of all devices connected to the Internet used IPv6 addresses; in 2014 3.5% of devices use IPv6
addresses. See Infographic, World IPv6 Launch, THE INTERNET SOCIETY,
http://www.worldipv6launch.org/infographic.
61
subject to Title II “only to the extent that it is engaged in providing telecommunications
services.”168 The Commission’s proposal to separate out the telecommunications component of
broadband service would run roughshod, accordingly, over Congress’ careful statutory
definitions of telecommunications service, on the one hand, and information service, on the
other. Furthermore, there is no principled distinction between the integrated offering of
broadband and other Internet-based information services, especially now that broadband
providers have begun to provide some of the same services as edge providers and content
delivery networks, as examined above.169
In an attempt to avoid this inevitable statutory problem, the Commission also requested
comment on whether to “separately identify and classify a broadband service that is furnished by
broadband providers’ [sic] to edge providers.”170 But this approach, which has been supported by
Mozilla and academic Tim Wu,171 is nothing new. As far back as the original Cable Broadband
proceeding, various parties urged the Commission to apply Computer II rules and require
broadband providers to offer a wholesale transmission service separate from the information
service aspects of broadband Internet access.172 But the Commission rejected that approach,173 as
168
47 U.S. C. § 153(44) (emphasis added).
169
Supra 7-9; infra 76-77.
170
See NPRM ¶ 151.
171
See Mozilla, Petition to Recognize Remote Delivery Services in Terminating Access Networks and Classify
Such Services as Telecommunications Services under Title II of the Communications Act, GN Docket Nos. 14-28,
10-127 & 09-191, at 4-5, 9 (May 5, 2014) (“Mozilla Petition”); Letter from Tim Wu and Tejas Narechania,
Columbia University to Marlene H. Dortch, Federal Communications Commission, Open Internet Remand, GN
Docket No. 14-28 (Apr. 14, 2014).
172
See Cable Broadband Order ¶ 42 .
173
See id. ¶ 43; see also id. ¶ 39 (“Cable modem service is not itself and does not include an offering of
telecommunications service . . . [Its] telecommunications component is not . . . separable from the data-processing
capabilities of the service.”); FCC Brand X Br. at 6 (“The Commission rejected arguments that the
telecommunications component of an information service should be isolated and viewed separately as an offering of
telecommunications.”) (citing Stevens Report ¶ 40).
62
did the Supreme Court when proponents renewed the argument there.174 And for good reason –as
explained below, such an approach would effectively amount to unbundling Internet access
service into its component parts, which would complicate service delivery, increase its cost, and
discourage consumer use of and provider investment in broadband service.
To the extent proponents now argue that the Commission should hypothesize a separate
transmission service offered to edge providers (basically, a virtual wholesale arrangement), the
argument is simply repackaging previously rejected arguments and mischaracterizing the
services that providers actually offer to the public. Broadband providers offer a single integrated
service to consumers that gives them not just access to their choice of content on the Internet, but
also security, the ability to back up data, email, cloud-based software services, and web hosting.
And to the extent that broadband providers enter into arrangements with differentiated economic
terms for specific edge providers, they are still offering the edge providers access to that same
service.175
It would blink reality to pretend that broadband providers offer edge providers a different
service than the broadband Internet access service that they actually provide and that is the
subject of this proceeding. As a result, the Commission could apply Title II in this framework
and overcome these metaphysical hurdles only by requiring broadband providers to unbundle
their service into its component parts. But that approach would merely resurrect all the problems
that have caused the Commission to reject it repeatedly since the Clinton administration:
unbundling would create prohibitive complexities in delivering separate services; customers
174
See Brand X, 545 U.S. at 999-1000 (describing how edge provider content sent to consumers is part of the
information service provided by broadband Internet access service).
175
See Cable Broadband Order ¶ 38 (“Accordingly, we find that cable modem service, an Internet access
service, is an information service. This is so regardless of whether subscribers use all of the functions provided as
part of the service, such as e-mail or web-hosting”).
63
would have to pay for both types of services, which would raise consumer costs; and both
problems would drive away consumers and providers from broadband service. As a result, all
edge providers would have to offer a variety of information service capabilities currently
bundled into broadband service to enable their services.
There is nothing “minimal” about this proposal, nor is it a “compromise,” as Mozilla
maintains.176 Even apart from the above threshold problems, it would be a dangerous course to
treat even just a component of broadband service – particularly one as ill-defined as the virtual
wholesale arrangement Mozilla describes – as a separate telecommunications service subject to
Title II. As with reclassification more generally, such an approach would require broadband
providers to charge edge providers for every bit transmitted over the network. The statute on its
face recognizes that “telecommunications services” are limited to those that are made available
to the public “for a fee,” so this approach provides no path to prohibiting paid arrangements
between edge providers and broadband providers and instead provides a firm statutory hook for
charges. Moreover, even if the Commission could somehow characterize a price of zero for the
theoretical edge provider services at issue as a “reasonable” fee – which it could not177 – how
this precedent could be translated in the international context is far from certain. In fact, other
countries have already displayed some interest in having termination charges paid by U.S.
Internet companies provide a larger portion of funding for those countries’ domestic broadband
networks. Add to this the risk that reclassifying one segment of broadband service would open
176
See Mozilla Petition, at 12.
177
Any effort by the Commission to set a price of zero for edge provider transmissions would only exacerbate
the Takings problems associated with compelled common carrier status, see infra at 67-69,and pose other questions
about the Commission’s power to set just and reasonable rates under Title II.
64
the door to classifying “essentially all” Internet-based services as telecommunications services
under Title II, and you have a prescription for policy disaster.178
In short, the Mozilla/Wu approach would lead to the same negative outcomes as broader
reclassification, including having serious international ramifications for large, U.S.-based
Internet companies,179 in addition to creating all sorts of problems related to unbundling
broadband Internet access service.
The FCC Lacks Statutory and Constitutional Authority to Compel Common Carriage.
Reclassification also would be unlawful because the Commission lacks authority to compel
common carriage. The Commission asks whether it should “apply Title II to broadband Internet
access service (or components thereof).”180 But this is not a decision that the Commission can
make at will in order to accomplish the desired policy result du jour. As the D.C. Circuit
established in NARUC I, “[a] particular system is a common carrier by virtue of its functions,
rather than because it is declared to be so” by the Commission.181 In other words, common
carrier status attaches only if the provider chooses to “undertake[] to carry for all people
indifferently,” meaning that it does not “make individualized decisions, in particular cases,
whether and on what terms to deal.”182 That is why common carriage made sense in the case of a
monopoly telephone world, but does not make sense for broadband providers.
178
Supra Section IV.D.
179
Id.
180
NPRM ¶ 149.
181
NARUC I, 525 F.2d at 644; see id. (rejecting “unfettered discretion in the Commission to confer or not
confer common carrier status on a given entity, depending upon the regulatory goals it seeks to achieve.”).
182
Id. at 641; see NARUC v. FCC, 533 F.2d 601, 608 (D.C. Cir. 1976) (“NARUC II”) (“[T]he primary sine qua
non” of the analysis is whether the carrier “holds himself to serve indifferently all potential users.”).
65
Without such voluntary undertaking to carry for the public, compelled common carriage
would constitute a government taking.183 Historically, such obligations have been permissible
only where carriers agreed to hold themselves out as willing to serve all in exchange for a
government-granted franchise with limited or no competition.184 In exchange for that
government guarantee, carriers were permitted to charge a rate that would provide an adequate
return; this assured return and restrictions on market entry by others allowed carriers to recoup
their investment.
None of that, however, applies to broadband. There is no monopoly and no government
restriction on entry.185 The market is competitive, and the government does not guarantee a
return. In fact, individualized service determinations have been integral to broadband providers’
experimentation with various services and have improved their ability to address consumer
demands.186 Under these circumstances, the Commission’s election, as a matter of pure
discretion, to apply Title II common carrier regulations to broadband providers would constitute
an unlawful government taking. Broadband providers such as Verizon possess vested property
rights in the physical infrastructure of the networks they built with billions of private investment
dollars. If the Commission were to force providers to dedicate those networks to the use of
others, that would constitute a government-mandated use of private property for public ends – a
183
As NARUC I recognized, early common carrier regulations were “challenged as deprivations of property
without due process.” NARUC I, 525 F.2d at 640.
184
See NARUC I, 525 F.2d at 642 (concept of common carrier “developed as a sort of quid pro quo whereby a
carrier was made to bear a special burden of care, in exchange for the privilege of soliciting the public’s business”);
AT&T Submarine Sys., Inc.; Application for a License to Land and Operate a Digital Submarine Cable System
Between St. Thomas and St. Croix in the U.S. Virgin Islands, Memorandum Opinion and Order, 13 FCC Rcd 21,585,
¶¶ 7-9 (1998) (the decision to impose common carriage depends on whether “the public interest . . . require[s] the
carrier to be legally compelled to serve the public indifferently” because the carrier “has sufficient market power”).
185
See NARUC I, 525 F.2d at 640; Verizon Commc’ns Inc. v. FCC, 535 U.S. 467, 477-89 (2002) (describing
how hallmarks of common carriage regulation developed to offset monopoly power in early history of the telephone
industry).
186
See supra at 18-19.
66
per se taking.187 Even if the Commission purported to provide compensation for that taking, that
would not solve the basic problem of the lack of any clear mandate from Congress to engage in
the taking in the first place.188 In any event, compelled common carrier status would constitute
a regulatory taking because such action would “interfere[] with [broadband providers’] distinct
investment-backed expectations,” Penn Cent. Trans. Co. v. City of New York, 438 U.S. 104, 124
(1978), to manage access to their networks and use them to provide the services customers want.
Compelled common carriage also would violate the First Amendment rights of broadband
Internet access providers, who use their platform to “engage in and transmit speech”189 – namely,
their own Internet services and others’ content services. Broadband networks are the modernday equivalent of the printing press. Those networks, no less than printing presses or newspaper
stands, cannot be conscripted for speech on the government’s terms.190 Nor can the government
discriminate against speakers by singling out broadband service providers (and their prospective
content partners) from Internet backbone services, content delivery networks, and other services
187
See Loretto v. Teleprompter Manhattan CATV, 458 U.S. 419 (1982); D. Lyons, Virtual Takings: The
Coming Fifth Amendment Challenge to Net Neutrality Regulation, 86 Notre Dame. L. Rev. 65, 93 (2011)
(explaining that in a common carriage regime edge providers would “receive an unlimited, continuous right of
access to broadband providers’ private property,” which “allows them to physically invade broadband networks with
their electronic signals and permanently occupy portions of network capacity”).
188
See Bell Atlantic Tel. Cos. v. FCC, 24 F.3d 1441, 1444-46 (D.C. Cir. 1994);see also Ramirez de Arellano v.
Weinberger, 745 F.2d 1500, 1510 (D.C. Cir. 1984) (en banc) (“When there is no authorization by an act of Congress
or the Constitution for the Executive to take private property, an effective taking by the Executive is unlawful
because it usurps Congress’s constitutionally granted powers of lawmaking and appropriation.”), overturned on
other grounds, 471 U.S. 1113 (1985).
189
Turner Broad. Sys., Inc. v. FCC, 512 U.S. 622, 636 (1994) (describing similar First Amendment rights of
cable programmers and cable operators).
190
See, e.g., Citizens United v. FEC, 558 U.S. 310, 441 (2010) (noting that the First Amendment “reject[s] the
premise that the Government has an interest in equalizing the relative ability of individuals and groups to have their
voices heard”) (internal citation omitted); Miami Herald Publ’g Co. v. Tornillo, 418 U.S. 241, 257 (1974)
(“Government-enforced right of access inescapably dampens the vigor and limits the variety of public debate.”)
(internal quotation omitted).
67
that have a similar ability to exercise editorial control over the delivery of content.191 These
constitutional concerns are further reason not to pursue reclassification.
Nor could the Commission avoid the above legal hurdles by adopting Congressman
Waxman’s suggestion and “proceed[ing] under Section 706 but us[ing] Title II as a ‘backstop
authority.’”192 The Commission cannot use Title II as a “backstop,” for that would require the
very definition of arbitrary and capricious decision-making. Effectively, that proposal would
have the Commission issue two orders: one finding that broadband service is an information
service; the other simultaneously finding just the opposite, namely, that broadband service is a
telecommunications service. Certainly Fox permits an agency to change its mind later, but the
agency cannot simultaneously reach mutually exclusive outcomes on the same issue. Moreover,
backstop authority only creates more regulatory uncertainty. In order to use Title II in this
fashion, the Commission would need to exercise forbearance with its Title II provisions. As
explained above, that approach, however, would cause a long and investment-deterring legal
battle over the scope of the Commission’s forbearance.193
Congressman Waxman’s second proposal – that the Commission reclassify broadband
but exercise forbearance so long as the Section 706 rules remain in effect – fares no better. In
addition to suffering from all the problems that plague his first proposal, it fundamentally misses
that the Commission cannot impose common carriage requirements at all until broadband
providers choose to hold themselves out to the public indiscriminately. The second approach,
therefore, adds nothing except regulatory rigmarole and additional uncertainty related to the
scope and operation of any proposed forbearance.
191
See Citizens United, 558 U.S. at 353 (“differential treatment cannot be squared with the First
Amendment”).
192
NPRM ¶ 150 (describing letter from Representative Henry Waxman).
193
See supra Section IV.A.
68
As the Commission recognized long ago, all of these substantial legal hurdles would
engender regulatory uncertainty that would “discourage investment in facilities” and spur
broadband providers to “raise their prices and postpone or forego plans to deploy new broadband
infrastructure.”194 In this regard, exercising forbearance would not help, as legal uncertainty
about the extent of that forbearance and its permanence could still “chill innovation.”195
***
At this juncture in the history of the development of broadband Internet service,
reclassification, either in whole or piece parts designed to accomplish the same regulatory ends,
would be a radical and risky reversal of a successful policy that has fostered innovation, growth,
and better satisfaction of consumer needs. It would fail to accomplish what its proponents want,
all the while endangering the entire Internet ecosystem and benefiting only established Internetplayers. And because reclassification would violate both the Telecommunications Act of 1996
and the Constitution, it would only create further legal uncertainty that would chill innovation
and investment. Reclassification under Title II ultimately would deprive consumers of the
benefits of not just today’s foreseeable innovations but tomorrow’s as-yet unimagined
innovations. Stuffing broadband into a Title II box would create not an open Internet but one
flash-frozen in 2014. This profoundly counter-productive policy is not the path forward.
V.
THE COMMISSION SHOULD REFRAIN FROM IMPOSING ANY RULES ON
INTERNET INTERCONNECTION ARRANGEMENTS, SPECIALIZED
SERVICES, OR ENTERPRISE OFFERINGS.
Finally, if the Commission does adopt new open Internet rules, it should not extend them
beyond the scope of the earlier rules and should maintain a focus on mass market Internet access
194
FCC Brand X Br. at 31; see Stevens Report ¶ 47 (“Such a presumption [in favor of Title II regulation of
broadband providers] would be inconsistent with the deregulatory and procompetitive goals of the 1996 Act.”).
195
Id. ¶ 47.
69
services offered over fixed broadband networks. The Commission should not impose any new
rules on Internet interconnection arrangements, specialized services, or enterprise offerings.
Rules designed for the specific context of last-mile Internet access would not be appropriate in
any of these areas.
A. There Is No Reason to Apply Open Internet Rules to Peering, Transit, and
Other Internet Interconnection Arrangements.
Given the unique circumstances presented by Internet interconnection – including an
intensely competitive marketplace and a long history of successful market-based interactions –
there is no basis to subject peering, transit, or other forms of Internet interconnection to rules
designed for the specific context of last-mile Internet access. Rather, the Commission should
adopt its tentative conclusion to exclude such arrangements from any new rules it may adopt.196
As Chairman Wheeler repeatedly has recognized, Internet interconnection issues are distinct
from the last-mile access issues involved in the “open Internet” debate.197
Internet traffic can reach an ISP in many ways. Throughout the Internet’s history,
content providers and their service providers have relied on commercially negotiated agreements
with backbone operators – network providers with extensive facilities used to carry Internet
traffic around the globe. These backbone providers themselves make arrangements with other
196
NPRM ¶ 59.
197
See, e.g., Statement of Chairman Tom Wheeler, Press Conference, June 13, 2014 (stating that net neutrality
issues and peering issues may be “cousins, but they’re not the same thing”); FCC Chairman Tom Wheeler: “A lot of
people seem to think the whole peering and interconnection topic is the same as net neutrality. It’s not, it’s a
different issue — it’s a cousin, maybe a sibling, but it is not the same issue.” Stacey Higginbotham, The NetflixComcast agreement isn’t a network neutrality violation…GIGAOM (Feb. 23, 2014 5:37 PM),
http://gigaom.com/2014/02/23/the-netflix-comcast-agreement-isnt-a-network-neutrality-violation-but-it-is-aproblem/ (quoting Jan. 28, 2014 CSPAN interview with Chairman Wheeler); Oversight of the Federal
Communications: Hearing Before the Comm’ns and Technology Subcomm. of the H. Comm. of Energy and
Commerce, 113th Cong. (May 20, 2014) (see Bryce Baschuk, FCC Chairman Defends Net Neutrality Proposal to
Skeptical Lawmakers, Bloomberg BNA (May 22, 2014 ), http://www.bna.com/fcc-chairman-defendsn17179890700/ (Chairman Wheeler states “peering… is ’a separate issue’”); Matthew Schwartz, Wheeler Promises
FCC Will Be ‘Nimble’ as Technology Evolves, COMMUNICATIONS DAILY, Feb. 11, 2014 (quoting Chairman Wheeler
statement that backbone interconnection “has been lumped into the ‘open Internet’ kind of issues, but I don't think it
really is [part of that set of issues]”).
70
backbone providers, and traffic carried between Internet endpoints often transits multiple
backbone networks. The commercial agreements between networks might create “peering”
relationships, in which networks interconnect directly and exchange traffic, or “transit”
relationships, in which one provider agrees to ensure that another provider’s traffic will reach its
destination, even if it must travel over the networks of additional other providers. Transit
agreements have always involved payment. Peering agreements may or may not involve
payment through the exchange of money: In some cases, the networks agree that it is simpler to
“call it even” and exchange traffic on a settlement-free basis without incurring the cost of
measuring it and making payments. In other cases – especially when the traffic exchanged
between networks is not roughly in balance – the provider receiving more traffic than it sends
will charge the other provider to peer with it, thus helping to compensate for the receiving
provider’s greater relative costs to handle the other network’s traffic.
Over time, players in the Internet ecosystem have created new and innovative
interconnection arrangements in response to changes in end users’ demands. Content delivery
networks developed to hasten traffic delivery, and now use geographically dispersed servers to
store (“cache”) content nearer to end users. CDNs, which allow traffic to reach end users more
quickly and with less latency or jitter than traffic travelling over a series of backbone networks,
have developed into a principal means for delivering video and other high-bandwidth content.
Companies such as Akamai, Limelight and Verizon’s EdgeCast sell CDN services to third-party
content providers, and numerous entities – most notably Google and Netflix – have constructed
their own CDNs. CDNs, in turn, often rely on transit providers to deliver their cached content to
ISPs.198
198
See Lerner Decl. ¶¶ 57-60.
71
As new business models have arisen, the Internet itself has shifted from a hierarchical
network featuring large Internet backbones interconnecting with smaller backbones and
(ultimately) the ISPs serving content providers and end users into a much more complex network
in which providers interconnect in a multitude of ways. In addition to relying on CDNs to
aggregate traffic and deliver them to ISPs (as Netflix used to do for all of its traffic), a content
provider can choose from among at least a dozen major providers of “transit” services whose
business models revolve around the delivery of traffic to ISPs, or can interconnect directly with
the ISP itself. Verizon’s network is one of the most interconnected in the world. Verizon has
hundreds of agreements involving the exchange of U.S. Internet traffic with Verizon’s last-mile
and backbone networks. These include, but are not limited to, arrangements such as Internet
access, transit, peering, colocation, hosting, and content distribution. The parties on the other
side of these agreements include Internet backbone providers, transit providers, ISPs, CDNs, and
edge providers. While Verizon’s recent interconnection deal with Netflix has drawn public
attention, other content providers that make up a significant percentage of overall traffic have for
years entered into similar arrangements to directly interconnect with our network. They have
generally found that such agreements are a successful and cost-effective way to provide highquality service to their own customers without the need to rely on Internet middlemen. The
breadth and variety of these agreements reflect that there are many ways to reach Verizon’s
customers and that Verizon is a cooperative, good-faith player in the interconnection market.
The need for flexibility – and the complexity of this sector – has increased over time as
many companies are assuming multiple roles in the Internet ecosystem. Verizon, for example, is
not only an ISP but also a backbone network providing both peering and transit service. Last
year, Verizon purchased EdgeCast, which operates a CDN that interconnects with other last-mile
72
ISPs around the globe. Likewise, Google, Netflix and many other large content providers now
control significant networks of their own, and often operate their own CDNs. The marketplace
has prompted other providers to adopt a similar range of roles to best meet customer needs.199
This diversity of roles and interconnection options has become critical to the Internet’s
functioning – without them, the Internet might still be optimized for text-based news sites and
blogs rather than for streaming massive volumes of HD content. Moreover, even amidst
burgeoning complexity, this system has functioned smoothly, and traffic has reached its
destination. But the robust ecosystem we enjoy today would not exist if the Commission had
adopted a regulatory approach to Internet interconnection rather than the market-driven approach
it chose. Providers and regulators alike would have been hard-pressed to predict the swift rise of
online video and other high-volume traffic, and rules premised on a pure peering/transit
framework could have brought the Internet to its knees as traffic patterns shifted and volumes
exploded. Fortunately, the Commission relied instead on a flexible commercial approach, which
permitted a diversity of interconnection arrangements and the rise of CDNs, and allowed
backbone providers to meet consumer demand by deploying new facilities where necessary.
These investments have, in turn, promoted competition, and prices for transit continue to fall –
creating a de facto limit on paid peering rates.200
199
For a general discussion of Internet architecture, see, e.g., Jonathan E. Nuechterlein and Philip J. Weiser,
DIGITAL CROSSROADS: TELECOMMUNICATIONS LAW AND POLICY IN THE INTERNET AGE 178-85 (2d ed. 2013). For a
discussion of the peering and transiting regime as it existed in 2000, see Michael Kende, Office of Plans and Policy
Working Paper, The Digital Handshake: Connecting Internet Backbones, Federal Communications Commission,
(Sept. 2000), http://transition.fcc.gov/Bureaus/OPP/working_papers/oppwp32.pdf.
200
In connection with a hearing before Congress, the founder and CEO of Cogent, a leading transit provider,
submitted written testimony stating that Cogent had reduced its prices by 22 percent per year over the past five
years. See Written Statement of Dave Schaeffer, Chairman and CE O, Cogent Communications Group, Inc.,
Hearing Before the Regulatory Reform, Commercial and Antitrust Law Subcomm. Of the H. Comm. of the
Judiciary, at 3 (May 8, 2014) , http://judiciary.house.gov/_cache/files/d89e8174-d014-4ade-8a0058c5b9350dd4/schaeffer-testimony.pdf.
73
So, too, it will be difficult – if not impossible – to predict what new arrangements will
arise to serve consumers’ and edge providers’ needs going forward, as usage patterns, content
offerings, and capacity levels continue to evolve. Under these circumstances, regulations
cabining the scope of permissible interconnection arrangements would undercut consumer
interests, distorting or impeding the Internet’s ability to serve consumers’ ever-changing needs.
Notwithstanding the long track record of success from the market-based approach to
Internet interconnection, Netflix and some of its transit providers continue to point the finger at
ISPs such as Verizon in an effort to expand the reach of net neutrality regulation beyond the last
mile and to obtain zero-cost, regulated access to ISPs’ networks. Netflix’s “strong net
neutrality” proposal is in fact a thinly disguised effort to jettison the long-standing, market-based
approach to Internet interconnection in favor of a regime in which content providers and their
transit carriers are entitled to free interconnection with an ISPs’ networks regardless of the costs
they impose or whether they have any obligation at all to carry traffic from that ISP in return.
This proposal would completely supplant the market-based approach that has encouraged
infrastructure investment and allowed the Internet to flourish even as usage has soared and
demands on networks have evolved. Netflix’s invocation of the “net neutrality” rubric is clever,
but inapt: Netflix merely seeks to avoid the costs that it and other content providers have always
incurred to deliver content and to place those costs – for the first time ever – on the ISP alone.
In pursuit of this agenda, Netflix has suggested that Verizon and other ISPs are causing
widespread congestion to the harm of consumers, but, as independent analysts have confirmed,
this is not so. Researchers at MIT led by David Clark recently issued a preliminary report
finding that there was not a “widespread congestion problem among the U.S. providers.”201 In
201
MIT Information Policy Project, Measuring Internet congestion: A preliminary report, at 2 (2014),
https://ipp.mit.edu/sites/default/files/documents/Congestion-handout-final.pdf.
74
fact, the report noted that most of the observed congestion related to Netflix traffic, and that “it
would appear that all parties are moving toward adequate resolution.”202 The MIT report also
noted that congestion on the Internet often resulted from “decisions by content providers [such as
Netflix] as to how to route content,” which can result in sudden congestion problems. Verizon’s
recent investigation into the cause of slow Netflix streaming experienced by one of its FiOS
customers confirmed the MIT report’s view.203 We found that the congested Netflix traffic was
caused by Netflix’s decision to route its traffic over a handful of transit providers who had not
made arrangements for connections that could handle Netflix’s traffic volumes, while the other
peering and transit providers and content providers interconnecting with Verizon’s network in
the customer’s area were not experiencing congestion. Thus, the solution to Netflix’s congestion
problems is for Netflix to negotiate for direct interconnection paths capable of handling its
unprecedented traffic volumes (as it has recently done with Comcast and Verizon) or to use other
network providers or CDN providers that have done so. There is simply no reason to reverse the
long-standing commercial approach to Internet interconnection, which has worked well and
evolved effectively over time with minimal bumps in the road for consumers. Indeed, the
commercial approach has worked here as well: As the recently negotiated arrangement between
Netflix and Verizon is implemented, the congestion problems caused by Netflix’s previous
decisions about how to route its flood of traffic should resolve. Notably, this pro-consumer
result transpired without regulatory involvement. The Commission should continue with its
approach to this area of the marketplace by encouraging the negotiation of such mutually
202
Id.
203
See David Young, Why is Netflix Buffering? Dispelling the Congestion Myth, VERIZON POLICY BLOG,
http://publicpolicy.verizon.com/blog/entry/why-is-netflix-buffering-dispelling-the-congestion-myth (last visited July
14, 2014).
75
beneficial arrangements for interconnection, and should continue to exclude such arrangements
from the scope of any open Internet rules.
B. There Is No Reason to Apply Open Internet Rules to Specialized Services.
In order to allow space for innovation, the Commission likewise should maintain its
approach of not applying open Internet rules to specialized services.204 As the Commission
previously recognized, specialized services “differ from broadband Internet access service and
may drive additional private investment in broadband networks and provide end users valued
services, supplementing the benefits of the open Internet.”205 Specialized services are by
definition distinct from the customer’s broadband Internet access service – they merely
supplement such service, increasing the range of options available to the consumer and
expanding consumer welfare. Based on this understanding of the pro-investment and proconsumer potential of specialized services, the Commission decided in 2010 to exempt
specialized services from the scope of its open Internet rules.206 It should reaffirm that choice
here.
As technology advances and turns concepts such as remote surgery, distance-learning,
and the Internet of Things into realities, the ability to offer specialized services could be critical
to promoting consumer interests and national policy priorities. Likewise, the promise of
specialized services will promote even greater investment in broadband infrastructure –
investment that benefits all users, whether or not they partake in such offerings themselves.
There is, moreover, no evidence suggesting that specialized services have “undermin[ed] or
threat[ened]” the open Internet, or that they will. Nor is there any reason to believe that
204
NPRM ¶ 59.
205
Open Internet Order ¶ 112.
206
See id. ¶ 113.
76
specialized services are limiting the growth or capacity available for broadband Internet access
service.207 Thus, there is no basis for the Commission to depart from its restrained approach.
The Commission also should refrain from further defining specialized services while they
are still in their infancy.208 The Open Internet Order’s definition – “other services [provisioned]
over the same last-mile connections used to provide broadband service”209 – is appropriately
broad, given that specialized services have yet to mature into a well-defined class of offerings.
During these early stages of their development, the Commission should allow specialized
services to evolve into a commonly understood category. In doing so, the Commission will
allow innovation to flourish, enabling industry players to develop new types of specialized
services that will “drive additional private investment in broadband networks and provide end
users valued services. . . .”210
C. There Is No Reason to Apply Open Internet Rules to Enterprise Services.
The Commission should continue to exclude enterprise services from its open Internet
rules.211 In 2010, the Commission observed that enterprise services, such as virtual private
network services, hosting, and data storage services, “typically are not mass market services
and/or do not provide the capability to transmit data to and receive data from all or substantially
all Internet endpoints.”212 As the NPRM recognizes, enterprise services “are typically offered to
207
See id. ¶ 114. In any event, providers must disclose whether specialized services affect available capacity
of broadband Internet access service in order to provide “effective disclosure” under the current transparency rule.
See id. ¶ 56.
208
209
210
NPRM ¶ 60.
Open Internet Order ¶ 7.
Id. ¶ 112.
211
See NPRM ¶ 58.
212
Open Internet Order ¶ 47.
77
larger organizations through customized or individually negotiated agreements.”213 Moreover,
the Commission has historically treated enterprise and mass-market broadband offerings
differently. There is no reason for the Commission to collapse this long-standing distinction
here. Indeed, imposing new terms that enterprise customers might not desire and have not
bargained for could render certain enterprise offerings uneconomic (e.g., by reducing investment
incentives and altering bargaining power) and undesirable (e.g., by disabling priority features
specifically sought by private IP enterprise customers). The Commission should refrain from
disturbing this highly competitive marketplace.
VI.
CONCLUSION
The consistent, light-touch approach to Internet regulation over the last two decades has
yielded enormous consumer benefits, and the Commission should maintain that course. Even as
the Commission considers how best to protect consumers and prevent harm to competition, it
should pursue a balanced approach that relies most heavily on informed consumer choices and
that allows flexibility to meet consumers’ evolving demands. The Commission should reject the
ultra-regulatory path of Title II “reclassification,” which would be a radical reversal that would
harm consumers and the continued development of the Internet, and would result in a protracted
period of investment-sapping uncertainty.
213
NPRM ¶ 58 (citing Open Internet Order ¶ 45).
78
Respectfully submitted,
/s/ William H. Johnson
Michael E. Glover
Of Counsel
William H. Johnson
Roy E. Litland
VERIZON
1320 North Courthouse Road
9th Floor
Arlington, VA 22201
(703) 351-3060
Attorneys for Verizon
and Verizon Wireless
Russell P. Hanser
WILKINSON BARKER KNAUER, LLP
2300 N St., NW
Suite 700
Washington, DC 20037
Helgi C. Walker
Kellam M. Conover*
GIBSON DUNN & CRUTCHER LLP
1050 Connecticut Ave., NW
Washington, DC 20036
*Admitted only in California; practicing under the
supervision of Principals of the Firm
July 15, 2014
79
EXHIBIT 1
Protecting and Promoting Consumer Benefits
Derived from the Internet
Declaration of Michael L. Katz
July 15, 2015
PROTECTING AND PROMOTING CONSUMER BENEFITS
DERIVED FROM THE INTERNET
Declaration of Michael L. Katz
July 15, 2014
CONTENTS
I.
INTRODUCTION AND OVERVIEW ............................................................................ 1
II.
THE COMMISSION’S FUNDAMENTAL APPROACH TO
REGULATING INTERNET ACCESS PROVIDERS SHOULD BE TO
PROMOTE AND PROTECT UNDISTORTED COMPETITION. ............................. 8
A.
FLEXIBILITY HAS BEEN KEY TO THE INTERNET’S SUCCESS IN
GENERATING CONSUMER BENEFITS. ........................................................................ 8
B.
PRIVATE INVESTMENT AND INNOVATION IN BROADBAND ACCESS
NETWORKS ARE VITAL TO CONSUMER WELFARE. ................................................. 12
III.
TRANSPARENCY REQUIREMENTS ........................................................................ 14
IV.
THE NO-BLOCKING RULE ........................................................................................ 19
V.
COMMERCIALLY REASONABLE PRACTICES .................................................... 23
A.
POTENTIAL GUIDING FACTORS ............................................................................... 24
B.
LIMITATIONS ON DIFFERENTIATED ARRANGEMENTS BETWEEN LASTMILE ACCESS PROVIDERS AND EDGE PROVIDERS WOULD VERY LIKELY
HARM COMPETITION AND CONSUMERS. ................................................................. 26
C.
SAFE HARBORS AND PRESUMPTIONS OF REASONABLENESS ................................... 31
1.
Non-Exclusive, Non-Affiliated Agreements ............................................... 31
2.
Sufficient Competition and Practices Already Used by Other
Broadband Internet Access Providers ....................................................... 32
3.
Two-Sided Pricing ..................................................................................... 33
VI.
IMPOSING NEW RULES ON MOBILE WIRELESS BROADBAND
INTERNET ACCESS PROVIDERS WOULD RISK DISTORTING
COMPETITION AND HARMING CONSUMERS. ................................................... 37
VII.
CONCLUSION ................................................................................................................ 40
VIII. APPENDIX: QUALIFICATIONS ................................................................................. 42
i
I.
INTRODUCTION AND OVERVIEW
1.
The Federal Communications Commission (Commission) has opened a proceeding in
which the Commission seeks comment on “[w]hat is the right public policy to ensure that the
Internet remains open?” 1 The Commission also seeks comments on several proposed rules.
2.
At the request of counsel for Verizon, I have conducted an economic analysis of the
likely effects of the Commission’s proposed rules on the consumer benefits derived from the
Internet. I analyze the proposed rules’ effects on consumer welfare because the maximization
of consumer welfare is the most fundamental public policy objective with respect to the
Internet. That analysis reveals that, if implemented incorrectly, the Commission’s proposed
rules would very likely harm competition, innovation, and investment, and, consequently,
would harm consumer welfare. However, if implemented in a flexible manner with
appropriate safe harbors and presumptions of reasonableness, the proposed rules could protect
consumer welfare.
3.
Briefly, my specific findings are the following:
x
Flexibility has been critical to the Internet’s success, and rules should preserve that
flexibility. The Internet ecosystem has been very flexible, to the great benefit of
consumers. As many have noted, that flexibility has allowed a wide range of new
applications and services to develop that make use of the Internet infrastructure,
including email, e-commerce, social networking, photo sharing, video conferencing,
1
Federal Communications Commission, Notice of Proposed Rulemaking, Protecting and
Promoting the Open Internet, GN Docket No. 14-28 (rel. May 15, 2014) (hereinafter, NPRM),
¶ 2.
and content streaming, to name just a few. Consumers have also benefited from other
forms of flexibility, including the flexibility with respect to business models and
business relationships. Prominent examples of new business models include content
delivery networks and a wide variety of advertiser-supported content services, such as
web search and video download services.
x
The Commission’s fundamental approach to promoting consumer benefits from the
Internet should be to promote and protect undistorted competition. No one knows
what will be the optimal forms of economic relationships between different
participants in the Internet ecosystem in the coming years. Rather than imposing a
particular structure and vision on market participants, the Commission should adopt
policies that promote undistorted competition. Doing so would best serve consumer
interests because it would allow those interests (as expressed through market forces) to
drive the services and applications that are commercially successful and, thus, are
offered to consumers. A policy approach of promoting undistorted competition would
allow the full expression of consumer desires and would allow Internet access and
edge providers to experiment with and implement a wide range of approaches as they
seek to discover which business models most successfully serve consumer interests.
x
Rules that restrict flexibility or take a one-size-fits-all approach risk distorting
competition and harming innovation, investment, and consumer welfare. To the
extent that the Commission’s rules lead to better-informed consumers and allow firms
to respond flexibly in innovative ways to consumers’ demands, these rules can
promote competition and consumer welfare. However, to the extent that the
2
Commission’s rules constitute de jure or de facto prohibitions of certain business
models, pricing strategies, or the full use of network management techniques, these
rules risk distorting competition and harming consumers. Consumer welfare is best
protected if the Commission allows broadband Internet access providers to manage
their networks and offer differentiated services or implement innovative pricing
strategies as long as those practices do not harm competition. Moreover, where
conduct is disallowed, public policy should do so only in response to specific
instances of identified harm, rather than imposing sweeping prohibitions that throw
out the good with the bad.
x
Transparency regarding service offerings and network management practices can
promote competition and consumer welfare by allowing consumers to make betterinformed choices, but ill-formed or arbitrarily applied rules can have adverse
unintended consequences and impose unnecessary costs. The Commission’s
transparency policy should: (a) generate information that aids decision making by
real-world consumers; (b) be flexible without creating excessive uncertainty; (c)
recognize the possibility of adverse unintended consequences; and (d) account for the
presence of existing consumer protection laws and policies. The Commission should
also recognize the roles of firms other than last-mile access providers in shaping end
users’ experiences. For example, congestion can be caused by the decisions of content
providers, transit providers, and content delivery networks concerning traffic routing,
network interconnection, and server capacity, among other things. The Commission’s
3
transparency rule should not require access providers to be “transparent” about
information or practices outside of their control.
x
In order to promote consumer welfare, the focus of any no-blocking rule should be on
anticompetitive blocking. Presumably, consumers would like access to whatever edge
providers they choose, and market forces generally have driven and will continue to
drive Internet access providers to offer services that do not block access to any lawful
edge provider’s service. Moreover, several access providers, including Verizon, have
publicly committed not to block or degrade lawful content regardless of its source. If
the Commission implements a no-blocking rule, it should specifically focus on
blocking that demonstrably harms competition. The no-blocking rule should also
reflect the facts that: (a) an overly broad interpretation of “blocking” can distort or
stifle competition and harm consumers by interfering with legitimate network
management functions or service features that give consumers greater options for their
Internet experience, and (b) the presence of a vigorous antitrust enforcement regime
attenuates the benefits of the Commission’s imposing a no-blocking rule.
x
Unless clear and meaningful safe harbors or presumptions are identified, a rule
prohibiting “commercially unreasonable” practices could distort and discourage
network management, innovative pricing, and other business practices that would
otherwise benefit consumers. Network management can facilitate more efficient use
of capacity and can protect consumers from harmful traffic and applications (e.g.,
spam or viruses that create zombie computers). However, unless properly formulated
and applied, the proposed rule would create an uncertain regulatory environment that
4
could discourage efficient network management. Similarly, unless properly
formulated and applied, the proposed rule could block or slow the emergence of new
pricing and service models that could otherwise benefit consumers by facilitating the
entry of new edge providers and/or promoting efficient, pro-consumer behavior by
incumbent edge providers.
— A per se prohibition of differentiated arrangements would limit flexibility and
innovation and, thus, very likely harm consumers without offering any offsetting
benefit. Arguments that differentiated arrangements harm competition and
consumers because some edge providers might gain competitive advantage over
others are based on a misunderstanding of the competitive process and how it
benefits consumers. Allowing an edge provider to make expenditures to improve
its service through a differentiated arrangement is a pro-competitive action that
benefits end users directly (by offering them a more-attractive service option) and
indirectly (by increasing the competitive pressures faced by rival edge providers).
Moreover, a differentiated access arrangement might be an important component
of a new entrant’s strategy for challenging well-established incumbents that
already purchased superior Internet access by building their own backbone
networks or making extensive use of content delivery networks (CDNs).
Flexibility to experiment with differentiated arrangements that are of interest to
edge providers and end users is thus a critical component of any regulatory regime,
and a ban on such arrangements could give rise to substantial harms in the future
as the Internet ecosystem continues to evolve. Moreover, a per se prohibition
5
would offer little or no incremental benefit over existing state and federal antitrust
policies of general applicability that already provide fundamental protections
should there be particular instances of differentiated arrangements that harm
competition.
— Two-sided pricing should be presumptively reasonable. Two-sided pricing models
(e.g., an edge provider’s helping to cover end users’ Internet access costs) can be
an innovative and valuable component of Internet access and edge providers’
competitive strategies that benefits consumers. The Commission should not
intentionally or unintentionally create barriers to such pricing by leaving it under a
cloud of regulatory uncertainty. Instead, the Commission should declare twosided pricing to be presumptively reasonable. The fundamental rationale for a
presumption of reasonableness is that an edge provider’s decision to pay all or part
of end users’ costs of gaining access to its services improves the value of its
services and does not degrade or reduce the value of the services offered by rival
edge providers. 2 In other words, it is competition on the merits.
x
Imposing new rules on mobile wireless broadband Internet access providers risks
distorting competition and harming consumers. Application of additional rules to
wireless broadband Internet access providers poses substantial risks of distorting
competition and harming consumers. This is so for at least three broad reasons. First,
2
For example, if the Commission implements a rule prohibiting broadband Internet access
providers from blocking edge providers from a baseline level of end-user access, that option
would remain open to an edge provider regardless of whether a rival edge provider chose to
pay all or part of end users’ costs of gaining access to its services.
6
the wireless industry very successfully offers a variety of mobile Internet access
services. Given the strength of competition, there is relatively little reason to believe
that pervasive regulatory intervention will improve market performance. Instead,
applying new rules to the wireless industry may distort competition to consumers’
detriment. Second, although network management and revenue-model innovation are
important to the successful operation of any communications network, they are
especially important for wireless networks given the capacity constraints they face
based on spectrum limitations. Third, Internet access providers offering both wireline
and wireless broadband services continue to invest heavily in their networks, but the
investment demands for wireless broadband service providers are particularly great as
they have to deal with spectrum scarcity and the need to deploy new technology to
meet the needs of growing numbers of subscribers and expanding mobile broadband
usage. Moreover, the value of investments in innovative network technologies, such
as 5G, that promise to deliver dramatically higher speeds to individual users are
particularly large. Avoiding inflexible rules that discourage this innovation and
investment will increase the degree to which mobile broadband Internet access
services provide an effective competitive alternative to fixed-line services.
4.
The remainder of this declaration explains these findings in greater depth and provides
details of the facts and analysis that led me to reach them.
7
II.
THE COMMISSION’S FUNDAMENTAL APPROACH TO
REGULATING INTERNET ACCESS PROVIDERS SHOULD BE
TO PROMOTE AND PROTECT UNDISTORTED COMPETITION.
5.
The central question of the design and implementation of Commission policy toward
broadband Internet access providers should be: What regulatory regime best promotes
consumer welfare? As I now discuss, the answer is: a regime that promotes and protects
undistorted competition. As I will also discuss, in order to promote consumer welfare, it is
essential that the regulatory regime not deny broadband Internet access providers the
flexibility and incentives needed to make the innovations and investments necessary to meet
consumers’ evolving demands.
6.
In recent decades, the Commission has overseen a fundamental shift in
telecommunications policy from an approach that created and regulated monopolies to one
that promotes competition and relies on market forces to “regulate” provider behavior. This
shift has given rise to tremendous consumer benefits in a wide range of services, including
long distance telephony, multichannel video distribution, and mobile wireless voice and data
services. With appropriate public policies in place, competition will continue to generate
consumer benefits in the Internet access marketplace in the form of lower prices, greater
variety, and higher product and service quality.
A.
7.
FLEXIBILITY HAS BEEN KEY TO THE INTERNET’S SUCCESS IN
GENERATING CONSUMER BENEFITS.
Public policies promote consumer welfare when they create an economic environment
in which firms have incentives to engage in investment and innovation that satisfy consumer
demands. Given the complexity, variety, and changing nature of consumer demands, this is
8
best done by promoting undistorted competition. Public policies that attempt to dictate the
course of market evolution are unlikely to serve consumer interests.
8.
Market experience demonstrates that consumers differ widely in the importance that
they attach to different characteristics of their Internet experience. Some customers prefer a
highly managed approach with strong security, while others prefer a “simple pipe.” Some
customers prefer to make use of bandwidth-intensive applications, such as streaming media,
while other customers prefer to use their service for e-mail and other applications that require
relatively little bandwidth.
9.
Well-informed consumers are the best judges of their own preferences. Given the
diversity of consumer preferences in the broadband user population, consumer welfare is
maximized when consumers are free to choose from among a range of different types of user
experiences. Of course, it is not efficient to offer every conceivable type of user experience—
at some point, the cost of additional variety outweighs the benefit. Market forces can better
determine the range of experiences to offer consumers than can regulators.
10.
In response to competitive pressures and consumer demands, providers of fixed-line
broadband Internet access offer a range of speeds, while mobile wireless broadband Internet
access providers offer a variety of speeds, access devices, and applications. Providers also
offer a variety of business models, with varying degrees of integration between content,
access devices, and access services.
11.
In a competitive marketplace, those companies that satisfy consumers’ needs and
desires earn greater financial returns than those that do not. Competition thus drives firms to
9
act to the benefit of consumers and can play an important role both in promoting innovation
and investment and in ensuring that the benefits of that innovation and investment accrue to
consumers. It follows that policies that protect competition serve to promote consumer
welfare. It also follows that policies that distort competition generally harm consumer
welfare.
12.
Public policy is very unlikely to serve consumer interests when it substitutes
regulatory mandates for broadband Internet access providers’ business judgments regarding
which products to offer consumers and what business models to pursue. One important
reason why relying on competitive market forces is superior to regulatory fiat is that
regulators almost inevitably lack the information necessary to determine which supplier
actions will maximize consumer welfare. Network management and new business models are
both prime examples. This conclusion is not a criticism of the Commission’s abilities but
rather a statement about the difficulty of the problem. No one can say with certainty what the
best approach is, including the firms themselves, which is why we observe a variety of
practices as firms seek to find the best approach for their particular circumstances. There may
not be a single approach that is optimal in all circumstances. And, even if there were a unique
optimal approach, at present no one may know for certain what it is. Instead, different entities
have different views and opinions based on different experiences, skills, information, and
analyses.
13.
Like the Commission, network operators face a problem that is too difficult to solve
with certainty. This is precisely why consumer welfare is best promoted by relying on
competitive market forces and why flexibility and experimentation are valuable. In the
10
absence of restrictive regulations, suppliers in a competitive market will experiment with a
diverse array of approaches, and those approaches that are most successful at creating
consumer value will prevail. Broadband Internet access providers that adopt inefficient
network-management practices, business or pricing models, or other strategies that consumers
find objectionable will face pressures to change and will suffer adverse commercial
consequences if they do not. In order to promote consumer welfare, the Commission should
not implement a rule that would hamper access providers’ abilities to experiment with new
business models or services, or to engage in undertake complex network management
practices, unless that conduct is specifically found to harm competition.
14.
Public policy is also very unlikely to serve consumer interests when it favors
some sectors of the Internet ecosystem over others, say by forcing one sector to abide
by rules designed to make another sector’s business models more profitable. Instead,
consumer interests are best served by public policies that promote undistorted
competition in all sectors of the Internet ecosystem.
15.
In summary, consumers are offered a wide range of experiences using both fixed-line
and mobile wireless Internet access. The wide array of choices available benefits consumers,
both by offering a range of options that satisfy varied consumer demand today and by
allowing for the testing of alternative approaches (including different pricing models) to see
which will be the most successful in meeting consumer demands in the future. To further the
goal of maximizing consumer welfare, the Commission should adopt policies that promote
and protect the competitive process, allowing consumer preferences to determine winners and
losers in a marketplace characterized by undistorted competition. No one knows the “best”
11
structure for the Internet. Indeed, given the heterogeneity of consumer preferences, it is very
unlikely that there is a single structure that is best for everyone. Because consumers have
different preferences and those preferences change over time, and because technology is
constantly evolving, it is important to continue to allow a variety of approaches to be taken
and for experimentation to take place.
B.
16.
PRIVATE INVESTMENT AND INNOVATION IN BROADBAND ACCESS
NETWORKS ARE VITAL TO CONSUMER WELFARE.
Consumers have benefited from tremendous innovation and investment both at the
edge and in the core of the Internet. 3 The parallel paths are not a coincidence. Internet
services are an example of a systems service or product—consumer benefits are generated
when several different components (e.g., network equipment and protocols, personal
computers or wireless handsets, and applications) are used together. Each of these
components interacts with, and depends upon, other components in the system. For example,
applications that make intensive use of bandwidth, such as streaming video, require networks
that are capable of providing the bandwidth, operating systems that are capable of processing
the data, and hardware that is capable of displaying the output. It should also be recognized
that the investment in core networks necessary to support edge innovation is not limited to
increased capacity. For example, AT&T had to make several investments in its network
3
For a discussion of wireless innovation and investment, see Michael L. Katz, “Public Policy
Principles for Promoting Efficient Wireless Innovation and Investment,” Attachment to
Comments of AT&T, Inc., Fostering Innovation and Investment in the Wireless
Communications Market; A National Broadband Plan For Our Future, GN Docket Nos. 09157, FCC 09-51, September 30, 2009, available at
http://www.att.com/Common/about_us/public_policy/fcc_wireless_noi/Paper-Katz.pdf, site
visited July 14, 2014 (hereinafter, Katz Innovation White Paper).
12
hardware and software to support the introduction of the Apple iPhone. 4 And IPv6 allows for
several improvements in core network performance that would enable new or improved
applications. 5
17.
Evolving consumer demands and technology are creating new opportunities that
require significant network investment to realize. New applications and new types of access
devices (e.g., smartphones and tablets), as well as increases in the numbers of users and
access devices, will continue to place increasing demands on broadband Internet access
networks. So, too, will the continuing growth in the streaming and downloading video over
the Internet, particularly high-definition video. 6 These new applications and video streaming
are expected to require significant amounts of additional bandwidth. 7
18.
Consumers will continue to enjoy the full potential benefits of services provided by
the Internet only if there is significant continuing investment in access networks. There is
widespread agreement that the vast majority of investment in innovation and facilities for U.S.
4
Katz Innovation White Paper, ¶ 50.
5
Ger van den Broek, “Next Generation Applications Working Group Report,” IPv6 Task Force,
February 4, 2001, available at http://www.eu.ipv6tf.org/PublicDocuments/IPv6TF-Apps.pdf,
site visited July 14, 2014, at 3.
6
For example, in the second half of 2013, Netflix and YouTube accounted for over 50 percent
of broadband traffic at peak times on wireline broadband networks. (Parks Associates, “New
Trends in Digital Delivery and CDNs,” 2014, at vi.) Moreover, video’s share of Internet
traffic is expected to grow from 57 percent in 2012 to 69 percent in 2017. (Id. at 9.)
7
For example, the Commission’s Broadband Speed Guide indicates that, at a minimum,
consumers need a download speed of 4 Mbps to adequately stream HD video content.
(“Broadband Speed Guide,” available at http://www.fcc.gov/guides/broadband-speed-guide,
site visited June 19, 2014). Similarly, Netflix recommends a connection of at least 5 Mbps for
HD video and 25 Mbps for streaming Ultra HD 4K quality video. (“Netflix Internet
Connection Speed Recommendations,” available at https://help.netflix.com/en/node/306, site
visited June 19, 2014.)
13
broadband Internet access will be made by private parties, who will be motivated by the
prospect of profits generated by those investments. All else equal, the greater the expected
financial return from a given level of investment, the greater are the incentives to undertake
that investment. Conversely, public policies that reduce the financial returns to investment
weaken private investment incentives. Thus, it is essential to consumer welfare that public
policies do not harm private investment incentives. These considerations are especially
important in the light of the fact that many investments in broadband networks involve large
sunk costs and highly uncertain returns.
III.
TRANSPARENCY REQUIREMENTS
19.
In the Open Internet Order, the Commission adopted the following rule: 8
A person engaged in the provision of broadband Internet access service shall
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 the use
of such services and for content, application, service, and device providers to
develop, market, and maintain Internet offerings.
The NPRM concludes that the rule leads to disclosures that: 9
(1) help end users make informed choices regarding the purchase and use of
broadband services and increase end users’ confidence in broadband providers’
practices; (2) ensure that edge providers have access to broadband providers’
network information necessary to develop innovative new applications and
services; and (3) inform the Internet community and the Commission about
broadband providers’ practices and conduct that could impact Internet
openness.
8
Federal Communications Commission, Report and Order, Preserving the Open Internet
Broadband Industry Practices, GN Docket No. 09-191 (rel. December 23, 2010) (hereinafter,
Open Internet Order), ¶ 54 [internal footnote omitted].
9
NPRM, ¶ 66.
14
20.
Transparency regarding available service offerings and network management practices
can promote competition and consumer welfare by allowing consumers to make betterinformed choices. In the other direction, competition can promote transparency (as firms are
driven by market forces to provide information to their customers) if consumers know what to
ask for (in other words, if consumers know what they don’t know). These facts suggest that
public policy can play a valuable role in promoting transparency and competition by
educating consumers with respect to what questions they should ask service providers. There
may also be instances in which competitive forces alone are insufficient to promote efficient
disclosure or to prevent fraud, so that public policy oversight is required.
21.
In order to promote competition and consumer welfare, the Commission’s
transparency policy should: (a) generate information that aids decision making by real-world
consumers; (b) be flexible with creating excessive uncertainty; (c) recognize the possibility of
adverse unintended consequences; and (d) account for the presence of existing consumer
protection laws and policies.
22.
There is a risk that some of the information the Commission seeks to make
available may be unintelligible to most end users. The Commission has tentatively
concluded that it “should require that broadband providers disclose meaningful
information regarding the source, location, timing, speed, packet loss, and duration of
network congestion.” 10 It is unclear how providers would disseminate such
information to end users that in fact would be meaningful. The Commission itself
10
NPRM, ¶ 83.
15
recognizes that “the manner in which providers display information to consumers can
have as much impact on consumer decisions as the information itself.” 11 It is unlikely
that the economic benefit of providing potentially esoteric network information to end
users would outweigh the economic costs to both providers and end users.
23.
Moreover, the Commission’s transparency rule should be properly cabined to
ensure that broadband Internet access providers are responsible for reporting only on
the sources of congestion for which they are responsible, while recognizing that the
decisions and practices of others in the Internet ecosystem likewise affect performance
and end users’ experiences. Although the NPRM asks whether broadband Internet
access providers should be required to make heightened disclosures concerning
congestion, the NPRM ignores the fact that congestion affecting end users’
experiences may be caused by the decisions of content providers, transit providers,
content delivery networks, or others about how to route traffic, interconnect with other
networks, or ensure adequate server capacity, among other things. To require
broadband Internet access providers to disclose performance information unrelated to
their own facilities or services is potentially misleading to consumers. The recent
dispute between Netflix and Verizon highlights this issue. Netflix has publicly blamed
Verizon for the poor performance that subscribers have experienced while streaming
Netflix, but evidence suggests that Netflix’s own network practices have played a
11
NPRM, ¶ 72 [internal footnotes omitted].
16
significant role. 12
24.
If not properly implemented, the transparency rule risks creating unnecessary
uncertainty and burdens, and could mislead or confuse consumers. The current transparency
rule, by design, allows a broadband Internet access provider considerable flexibility in
satisfying the requirements. This flexibility is an important virtue because there are many
forms that network management practices and disclosure of those practices could take.
Moreover, some networks have frequent, temporary changes in network performance due to
maintenance, damage issues, or spikes in user demand. Providing consumers with real-time
updates is useful in some circumstances, but some measure of average performance may be a
sounder basis on which to choose among access providers. The complexity of these issues
increases the likelihood of adverse unintended consequences if the Commission were to try to
write detailed, prescriptive regulations.
25.
Although the general nature of the proposed rule creates flexibility, it also risks
creating uncertainty that could distort access provider behavior or lead to needless
misunderstandings that trigger costly administrative proceedings. It is my
understanding that, to date, the Commission has done a good job balancing the value
12
Dan Rayburn, “New Data Questions Netflix’s Assertion That ISPs Are At Fault For Poor
Quality,” Streaming Media Blog, available at
http://blog.streamingmedia.com/2014/06/netflix-isp-newdata.html, site visited July 10, 2014
(“When Netflix took over the routing controls for their video traffic with their own CDN Open
Connect, customer performance began to suffer as highlighted in Netflix’s own data that they
shared with the Washington Post.”); Ben Popper, “The war of words continues: Verizon says
Netflix is the one causing internet congestion,” The Verge, available at
http://www.theverge.com/2014/7/10/5888239/verizon-netflix-congestion, site visited July 11,
2014 (“A study by Sandvine found that Netflix was choosing over-utilized transit options for
its data when it had clear alternatives, a decision that resulted in performance issues.”)
17
of flexibility and the value of certainty. It should continue this approach rather than
attempt to dictate precise forms of disclosure.
26.
It is well documented that even well-intentioned regulations can impose
significant costs and often have harmful unintended consequences. 13 Given that the
proposed rules would inevitably have costs, it is important to establish that the rules
would have significant benefits that would outweigh these costs. In particular, a
policy approach that seeks to maximize consumer welfare would not impose new rules
before demonstrating that there is a need for such rules to supplement existing ones.
27.
Even where public policies mandating disclosure are potentially valuable, it is not
evident that sector-specific rules are necessary and—if they are—whether the Commission
should be the agency to impose them. There already exists a range of state and federal laws
and public policies of general applicability that provide fundamental consumer protections.
Additional, sector-specific rules may be redundant or may conflict with existing rules. And
13
Examples of unintended harmful effects from regulation include:
x
Rate of return regulation, from which regulators have largely moved away because of
concerns about inefficient production, overcapitalization, and negative effects on quality
and new service offerings. (Dennis Carlton and Jeffrey Perloff, Modern Industrial
Organization (Fourth Edition), 2005, at 707-714.)
x
The stringent conditions intended to promote public safety radio that were attached to the
D Block in the 700 MHz auction. The bids did not reach the reserve price, and the
Commission instituted a new proceeding to modify the requirements and modify the
reserve price. (Federal Communications Commission, Second Further Notice of Proposed
Rulemaking, WT Docket No. 06-150, May 14, 2008.) Consequently, the conditions did
not promote public safety radio and the failure to auction the spectrum rights led to delays
in the introduction of services using this spectrum.
x
Food labeling has had little success in changing diets. For example, reduced consumption
of one high-fat food is frequently offset by increased consumption of another high-fat
food. (See Omri Ben-Shahar and Carl E. Schneider (2011) “The Failure of Mandated
Disclosure,” University of Pennsylvania Law Review, 159 (3): 647-749, at 677.)
18
an advantage of economy-wide rules is that a well-developed body of case law and precedent
has already been developed. This established regime has the advantage of being relatively
flexible while at the same time providing a less uncertain environment for investors and
innovators than would a new set of sector-specific rules. For example, the Federal Trade
Commission has extensive experience with consumer protection policy, and private-sector
firms also have extensive experience complying with the Federal Trade Commission’s
policies.
IV.
THE NO-BLOCKING RULE
28.
The Commission tentatively concludes that it should adopt the following rule: 14
A person engaged in the provision of fixed broadband Internet access service,
insofar as such person is so engaged, shall not block lawful content,
applications, services, or nonharmful devices, subject to reasonable network
management.
A person engaged in the provision of mobile broadband Internet access
service, insofar as such person is so engaged, shall not block consumers from
accessing lawful websites, subject to reasonable network management; nor
shall such person block applications that compete with the provider’s voice or
video telephony services, subject to reasonable network management.
29.
For the reasons described in Section II above, the focus of this rule should be on
preventing practices that harm competition. A public policy that prevents broadband Internet
access providers from using blocking to harm competition can serve consumer interests.
Indeed, such a policy already is in force in the form of antitrust policy: when it can be
demonstrated to harm competition, foreclosure is an antitrust violation. It should be
14
NPRM, ¶ 94 [internal footnotes omitted].
19
recognized that the presence of a vigorous antitrust enforcement regime attenuates the
benefits of the Commission’s imposing a no-blocking rule.
30.
As some proponents of network neutrality regulation have argued, there are conditions
under which an integrated supplier with significant market power will find it profitable to use
that market power to exclude rival edge providers, harming competition and consumers.
However, it is far from a foregone conclusion that such an integrated service provider will
engage in such exclusion 15 or that instances in which a vertically integrated supplier appears
to favor its own applications or content are anticompetitive. 16 I am unaware of any evidence
indicating that there was a widespread problem of anticompetitive foreclosure by broadband
Internet access providers prior to imposition of the anti-blocking requirements of the Open
Internet Order. 17 I am similarly unaware of any evidence indicating that current public
policies, including antitrust enforcement, are not up to the task of preventing anticompetitive
15
Joseph Farrell and Phil Weiser (2003), “Modularity, Vertical Integration, and Open Access
Policies: Towards a Convergence of Antitrust and Regulation in the Internet Age,” Harvard
Journal of Law and Technology, 17(1), available at
http://jolt.law.harvard.edu/articles/pdf/v17/17HarvJLTech085.pdf, site visited July 14, 2014.
16
Instead, such favoritism may actually reflect efficiency benefits of greater coordination
facilitated by integration that result in higher levels of consumer welfare.
17
47 C.F.R. § 8.5.
The NPRM preceding the Open Internet Order identified only two incidents in which the
Commission perceived the need to take action against an Internet service provider’s blocking
or degrading of traffic. (Federal Communications Commission, Notice of Proposed
Rulemaking, Preserving the Open Internet Broadband Industry Practices, GN Docket No. 09191, WC Docket No. 07-52 (rel. October 22, 2009), ¶¶ 32, 36, and 50. Moreover, in 2007, a
staff report of the Federal Trade Commission stated that “to date we are unaware of any
significant market failure or demonstrated consumer harm from conduct by broadband
providers.” (Federal Trade Commission, “Broadband Connectivity Competition Policy,” Staff
Report, June 2007, available at http://www.ftc.gov/reports/broadband/v070000report.pdf, site
visited July 7, 2014, at 11.)
20
foreclosure. Lastly, major service providers have publicly committed not to block access to
any legal content or applications that their customers seek to access. 18 Hence, there is no
reason to expect that additional regulations targeted specifically at such practices would
significantly promote competition and consumer welfare.
31.
It should also be recognized that an overly broad interpretation of “blocking” can
distort or stifle competition and harm consumers. That is one of the central reasons why
antitrust enforcement requires evidence that the challenged conduct harms competition and,
thus, consumers. It is also a reason why the Commission should create one or more safe
harbors or presumptions that define practices presumed reasonable and that do not constitute
blocking. The NPRM addresses this issue by proposing to allow “individualized bargaining”
for levels of service above a baseline level of Internet access. 19 The NPRM also notes that: 20
One way to define a minimum level of access is as a requirement that
broadband providers apply no less than a “best effort” standard to deliver
traffic to end users. For any particular type of Internet traffic, best-effort
delivery would represent the “typical” level of service for that type of traffic—
in effect, routing traffic according to the “traditional” architecture of the
Internet. Broadband providers would be free to negotiate “better than typical”
delivery with edge providers, and would be prohibited (subject to reasonable
network management) from delivering “worse than typical” service in the form
18
See, for example, Verizon, “Verizon’s Commitment to Our Broadband Internet Access
Customers: Our Customers Get Everything the Open Internet Has to Offer,” available at
http://responsibility.verizon.com/broadband-commitment/, site visited July 10, 2014 (“On any
of our Internet access services, wireline or wireless, you and other users of our service can
access and use the legal content, applications, and services of your choice, regardless of their
source.”) and Josh Lowensohn, “Comcast, Verizon, and others promise net neutrality ruling
won’t hurt customers,” The Verge, available at
http://www.theverge.com/2014/1/14/5309268/comcast-verizon-and-others-promise-netneutrality-ruling-wont-hurt, site visited July 9, 2014.
19
NPRM, ¶ 95.
20
NPRM, ¶ 102 [internal footnotes omitted].
21
of degradation or outright blocking.
32.
A carefully crafted rule that ensures that traffic will not be blocked or degraded
over an end user’s best-effort Internet access service would provide assurances that
end users could access the content and applications that they desire and that edge
providers would continue to have a path to reach end users. That said, the
Commission’s proposal to define a specific minimum level of access based on the
“typical” level of service for a given type of traffic is problematic. For instance,
typical cannot be defined as the mean or average level because it is a simple fact of
arithmetic that—unless all traffic is treated identically—some traffic has to receive
below-average treatment.
33.
From the perspective of protecting competition and promoting consumer
welfare, a better approach than setting a hard-and-fast minimum based on problematic
interpretations of what is typical would be to establish a zone of reasonableness (rather
than a specific minimum) and to do so by relying on market metrics: specifically, end
users’ broadband Internet access choices. As long as an access provider does not
engage in source-based blocking or degradation of traffic relative to the level of besteffort Internet access service an end user reasonably expects based on the service he or
she has purchased, the no-blocking rule should allow arrangements offering service
above the baseline selected by the end user.
22
34.
As the NPRM acknowledges, there are important differences between fixed and
mobile broadband Internet access providers. 21 As a result of these differences, there are
heightened concerns with respect to applying the no-blocking rule to mobile broadband
Internet access providers. The Commission proposes to “prohibit mobile broadband providers
from blocking lawful web content as well as applications that compete with the mobile
broadband providers’ own voice or video telephony services, subject to reasonable network
management.” 22 The Commission also 23
seek[s] comment on whether it would serve the public interest to expand the
rule’s scope to include reasonable access to all applications that compete with
the mobile broadband Internet access provider’s other services, not just those
that compete with voice or video telephony services, subject to reasonable
network management practices.
The benefits of such an extension would be low and the risks of harm high given the lack of
evidence of a widespread problem today, the fact that the United States has vigorous antitrust
enforcement, and the characteristics of the mobile wireless industry described in Section VI
below.
V.
COMMERCIALLY REASONABLE PRACTICES
35.
The Commission tentatively concludes it “should adopt a rule requiring broadband
providers to use ‘commercially reasonable’ practices in the provision of broadband Internet
access service.” 24
21
NPRM, ¶ 140.
22
NPRM, ¶ 105.
23
NPRM, ¶ 106.
24
NPRM, ¶ 116.
23
36.
Unless properly formulated and applied, such a rule could discourage the use of
efficient network management and block or slow the emergence of new pricing and service
models that could otherwise benefit consumers. Such arrangements could facilitate the entry
of new edge providers and promote efficient, pro-consumer behavior by incumbent edge
providers. Because no one has the ability to predict what will be the best network
management practices and pricing and service models in the future, it is important that the
Commission’s rule be flexible and remain focused on addressing only those practices that
demonstrably harm competition. A case-by-case (or rule-of-reason) approach can offer that
flexibility. At the same time, such an approach risks creating uncertainty that could distort
provider behavior or trigger costly administrative proceedings. Contrary to the NPRM’s
assertion, a rule-of-reason, approach does not provide certainty25 unless either: (a) there is a
substantial record of adjudication, or (b) there are realistic safe harbors. The approach is
workable, however, if the Commission develops a set of well-constructed safe harbors or
presumptions that are reasonably available to broadband Internet access providers.
A.
37.
POTENTIAL GUIDING FACTORS
NPRM identifies several factors as potential guideposts. 26 The Commission proposes
the impact on present and future competition and the impact on consumers as two, distinct
25
NPRM, ¶ 136.
26
In addition to the factors discussed below, the Commission “propose[s] to adopt a factor or
factors in applying the commercially reasonable standard that assess the impact of broadband
provider practices on free exercise of speech and civic engagement” on the grounds that free
speech promotes broadband investment and deployment. (NPRM, ¶ 131.) However important
is the objective of promoting free speech, it is difficult, if not impossible, to see any genuine
link between free speech and the concept of “commercially reasonable.” Instead, adopting
24
guiding factors. 27 In fact, these two factors can and should be combined. Consideration of
the impact on consumers should be subsumed within consideration of the impact on
competition because the best way for the Commission’s rules to promote consumer welfare is
for the rules to promote competition. More specifically, consumer welfare is best protected if
the Commission allows broadband Internet access service providers to manage their networks
and—if they so choose—offer differentiated services or implement sophisticated pricing
strategies as long as those practices do not harm competition.
38.
In assessing whether a practice harms competition, it is essential not to confuse harm
to competitors with harm to competition. 28 The Commission has stated “We believe that
consumers of broadband access service should have the ability to exercise meaningful
choices.” 29 Those choices should include services offered by edge providers that have chosen
to enter arrangements for differential treatment in order to offer more desirable services to
their customers. Such options will benefit consumers, even as they reduce rival edge
providers’ economic welfare by increasing the competitive pressure they face.
39.
The NPRM also raises technical characteristics as a potential guiding factor. 30 The
Commission is correct to be concerned that application of any reasonableness rule should be
sensitive to the technical characteristics of the network involved. It should be recognized,
this guideline would amount to imposition of a public-interest or universal-service obligation
of the sort imposed on regulated monopoly, common carriers.
27
NPRM, ¶¶ 124 and 129.
28
It is also important to keep in mind the existence of current antitrust enforcement, which limits
potential harms from firm misconduct.
29
NPRM, ¶ 130.
30
NPRM, ¶ 132.
25
however, that there is a potential tension between the static and dynamic application of such a
factor. On the one hand, allowing less-capable networks more regulatory leeway could
promote efficiency to the extent that such networks are less able to adapt to the regulatory
requirements. On the other hand, such a policy acts as a tax on investment in network
capabilities when such investment triggers more stringent regulatory requirements.
40.
Lastly, the NPRM also raises industry practices as a potential guiding factor. 31 It is a
sensible approach to presume that widely adopted practices are reasonable, especially when
agreed to through multi-stakeholder forums or previously approved by the Commission.
B.
41.
LIMITATIONS ON DIFFERENTIATED ARRANGEMENTS BETWEEN
LAST-MILE ACCESS PROVIDERS AND EDGE PROVIDERS WOULD
VERY LIKELY HARM COMPETITION AND CONSUMERS.
The Commission seeks comment on whether it should permit broadband Internet
access providers to enter into differentiated arrangements with edge providers, or,
alternatively, restrict or prohibit some or all types of differentiated arrangements. 32 The
Commission also asks a series of questions regarding broadband providers’ incentives and
economic ability to engage in certain conduct: 33
To what extent do broadband providers today have economic incentives and
mechanisms to block or disadvantage a particular edge provider or class of
edge providers? To what extent do vertically integrated providers have
particularized incentives to discriminate—on price, quality, or other bases—in
favor of affiliated products? What are broadband providers’ incentives to
increase revenues by charging edge providers for access or prioritized access to
the broadband provider’s end users?
31
NPRM, ¶ 134.
32
NPRM, ¶¶ 89, 90, 111, and 138.
33
NPRM, ¶ 44.
26
42.
Before considering the merits of permitting, limiting, or banning differentiated
arrangements between last-mile Internet access providers and edge providers for differential
treatment, it is worth commenting on per se regulatory rules generally. Given the continuing
evolution of the market and the value of experimentation and flexibility, it is difficult—if not
impossible—to conceive of circumstances in which a rebuttable presumption of harm would
not be preferable to a per se prohibition of a given practice.
43.
Applications differ greatly in the requirements they place on a network. 34 Some
applications require little bandwidth and are not time sensitive, such as e-mail. Other
applications require significant bandwidth but are not time sensitive, such as downloading
software or (non-streaming) video. Still other applications are time sensitive and require
significant bandwidth, such as high-quality videoconferencing. The speeds applications are
projected to require in the near future range from a few kilobits per second for VoIP or music,
to several megabits per second for streaming television or games, and up to twenty-five
megabits per second for the latest video streaming resolution, Ultra HD 4K. 35 In the light of
these differences among applications, it should not be surprising that the current Internet
34
Applications also differ greatly in terms of their sensitivity to various quality of service
metrics, including delay, jitter, throughout, and packet loss. (See, for example, Broadband
Internet Technical Advisory Group, “Real-time Network Management of Internet
Congestion,” October 2013, § 3.5.)
35
The Commission’s Broadband Speed Guide indicates that consumers need a download speed
of only 0.5 Mbps for VoIP, but need 4 Mbps for HD video streaming. (“Federal
Communications Commission Broadband Speed Guide,” available at
http://www.fcc.gov/guides/broadband-speed-guide, site visited June 19, 2014.) Similarly,
Netflix recommends a connection speed of at least 5 Mbps for HD video and 25 Mbps for
Ultra HD 4K quality video. (“Netflix Internet Connection Speed Recommendations,”
available at https://help.netflix.com/en/node/306, site visited June 19, 2014.)
27
architecture favors some applications over others. More important, the diversity of needs and
preferences implies that variety, differentiation, and experimentation all can be valuable.
44.
It is commonplace for firms in the Internet ecosystem other than access providers to
offer multiple versions of their products and to do so under a variety of different business
models. For example, Hulu offers a baseline level of access to programming on an advertisersupported basis while Hulu Plus offers a wider range of access devices, greater content
variety, and higher video quality but requires end users to pay subscription fees. 36
45.
It is also common for edge providers to enter into arrangements to differentiate
treatment of their traffic in parts of the Internet other than the last mile. For instance, content
providers already purchase differential treatment through self-supply or third-party CDNs,
such as Akamai, Limelight and Tata/BitGravity, which sell services that improve the speed
and quality at which their customers’ content can be accessed by end users. One way in
which a CDN can improve performance is by maintaining a large number of diverse servers
that connect to the Internet near end users’ locations, thereby reducing the number of hops
between an edge provider’s content and end users. Similarly, large content providers, such as
Google, Microsoft, and Yahoo, operate private backbone networks that connect to the Internet
and allow those providers to enjoy higher quality transport of their packets than that available
on the public Internet. 37
36
Hulu, Frequently Asked Questions, available at http://www.hulu.com/plus, site visited July 8,
2014.
37
Drew Fitzgerald and Spencer E. Ante, “Tech Firms Push to Control Web’s Pipes,” The Wall
Street Journal (Online), available at
http://online.wsj.com/news/articles/SB10001424052702304173704579262361885883936, site
28
46.
The Commission should prohibit only differentiated arrangements between edge
providers and last-mile broadband Internet access providers that demonstrably harm
competition, rather than broadly prohibiting practices that could well enhance competition
and increase consumer welfare in a variety of contexts. For example, an edge provider might
want to enter into a differentiated access arrangement as part of a disruptive strategy that
could benefit end users by creating a new option for them and increasing the competitive
pressures faced by rival edge providers. It is also essential that the regulation be designed
taking into account the costs—including the likely costs of unintended consequences—and
the presence of existing antitrust and regulatory policies.
47.
Arguments against differentiated arrangements confuse harm to specific competitors
with harm to competition. 38 Allowing an edge provider to enter into differentiated
arrangements to improve its service is a pro-competitive action. For example, such an
arrangement might be an important component of a new entrant’s strategy for challenging
well-established incumbents, such as Google, Netflix, and Facebook, that already prioritize
traffic on their own networks. Precisely because it increases competitive pressures on rival
edge providers, such expenditures harm those rivals. To assert that such harm to competitors
constitutes harm to competition is to misunderstand completely the competitive process and
how it generates consumer benefits. Imposition of sweeping restrictions or prohibition of
visited July 10, 2014. See also, Craig Labovitz, “How Big is Google?” Arbor Networks,
available at http://www.arbornetworks.com/asert/2010/03/how-big-is-google/, site visited July
10, 2014.
38
Indeed, this confusion by opponents applies to differential treatment of edge providers by
Internet access providers generally (e.g., sponsored data and two-sided pricing).
29
differentiated arrangements thus risks becoming the Robinson-Patman Act for the Internet age
in that it is an anticompetitive policy masquerading as a pro-competitive one.
48.
Another way to see the flaw in the arguments calling for sweeping restrictions on—or
a prohibition of—differentiated agreements is to trace out the implications of those arguments
in other contexts. For example, applying the logic of opponents to differentiated agreements,
a restaurant would not be allowed to buy and serve more expensive, organic poultry on the
grounds that buying a superior input would disadvantage other restaurants that could not
afford to pay for organic poultry given their market positions. The logic of opponents to
differentiated arrangements could be used to argue for pervasive regulation of many
information and communication technology firms. For example, one could argue that use of
Microsoft Office could be very valuable to small firms but some of those firms find paying
for the software to be financially burdensome. Should Microsoft be forced to provide Office
for free in order to avoid reducing innovation by Office users and to ensure that there is no
“discrimination” against smaller companies that are less able to pay for software licenses than
are large companies? Should Microsoft have to provide Office for free given that its marginal
costs are near zero and selling Office creates a virtuous cycle that promotes the sales of
Windows?
49.
Similar arguments could be made with respect to Google’s web search services. The
treatment by Google search is arguably more important to the success of many edge providers
than are the precise characteristics of how that traffic is delivered over the last mile. By the
logic of many arguments against differentiated arrangements, Google should be prevented
from selling advertisements, which amount to differentiated search, on the grounds that doing
30
so disfavors firms that might not be able to “afford” such ads. The appropriate conclusion is
not that regulation should be extended to Google’s web search service. Rather, the
appropriate conclusion is that the approach advocated by many opponents of differentiated
arrangements would find a need for regulation even where there is none.
C.
50.
SAFE HARBORS AND PRESUMPTIONS OF REASONABLENESS
The NPRM identifies possible safe harbors. 39 These would play an important role in
providing a degree of certainty to industry participants while allowing the Commission to
pursue a flexible, case-by-case approach toward practices outside of the safe harbors. In
addition to absolute safe harbors, the Commission should also create rebuttable presumptions
of reasonableness for certain practices unlikely to harm competition. Other parties and/or the
Commission could seek to rebut the presumption of reasonableness on a case-by-case basis,
but the evidentiary burden would be on the party seeking to have the practice or conduct
determined to be unreasonable.
1.
51.
Non-Exclusive, Non-Affiliated Agreements
The Commission is considering creating a safe harbor for non-exclusive deals offered
to non-affiliated entities. 40 This approach is a sound one. The key point to recognize is that
the focus of the rule should be on potential harm to competition and whether certain practices
harm competition by differentially treating competing edge providers. If a deal is offered on a
39
NPRM, § III.E.4. Among the safe harbors, the Commission has tentatively concluded to
continue its approach of 2010 by excluding mobile broadband providers from the no
commercially unreasonable practices rule. (NPRM, ¶¶ 62 and 140.) For the reasons discussed
in Section VI below, this exclusion would promote undistorted competition and benefit
consumers.
40
NPRM, ¶ 141.
31
non-exclusive basis, then other edge providers can avail themselves of that deal if they believe
it will allow them to better differentiate their services or in some other way compete more
effectively. The non-exclusivity would ensure that the Internet access provider could not
“leverage” any market power that it might possess to harm competition in the relevant market
for edge provider services.
2.
52.
Sufficient Competition and Practices Already Used by Other
Broadband Internet Access Providers
In areas where consumers can choose among competing broadband Internet access
providers—particularly if more than one provider is offering very-high-speed access
services—there is less concern that an access provider will adopt policies that harm
competition among edge providers because attempts to do so may trigger consumers to switch
access providers. In addition to having less concern about the competitive effects of
differentiated arrangements, there is greater concern that regulation will impose costs by
reducing flexibility and discouraging experimentation. Competing broadband Internet access
providers may experiment with various ways to differentiate themselves from one another in
order to gain competitive advantage. The greater the degree of competition, the less concern
there will be that a given provider’s practices harm competition and consumers.
53.
The Commission should also consider adopting a presumption of reasonableness
where a firm can demonstrate that its practice has been undertaken by one or more firms
facing sufficient competition (possibly in geographic markets other than the one in which the
firm whose practice is at issue competes) or where the Commission has previously found that
practice to be commercially reasonable when used by another Internet access provider.
32
3.
54.
Two-Sided Pricing
Internet access providers operate two-sided platforms that bring together end users and
a wide variety of edge providers offering content and applications. It is commonplace for
platforms to charge users on both sides for bringing them together. For example, many
newspapers and magazines charge both their advertisers and subscribers. And end users
clearly can benefit when edge providers pay for connections. For example, Amazon promotes
the simplicity of its bundled offering, the Kindle Paperwhite reader, by stating that “3G
wireless connectivity is automatic (with no monthly fees or annual contracts).” 41 Similarly,
“Amazon pays for Kindle DX's wireless connectivity so [the Kindle user] won't see a wireless
bill.” 42 Moreover, many automobile manufacturers now offer telematics systems providing a
variety of security, navigation, and emergency services through mobile wireless connections
using a variety of pricing models including those that bundle the cost of the mobile wireless
connection in the price of the telematics service. 43 As these nascent industries continues to
develop, new business models will almost certainly be deployed involving varying degrees of
bundling and integration among applications, access devices, and access services.
41
Amazon, “Connect to 3G,” available at
http://www.amazon.com/gp/help/customer/display.html?nodeId=201301640, site visited June
19, 2014.
42
Amazon, “Wireless, Whispernet and Whispersync,” available at
http://www.amazon.com/gp/help/customer/display.html/ref=help_search_12?ie=UTF8&nodeId=200375910&qid=1261384011&sr=1-2#cost, site visited June 6, 2014.
43
See, for example, Antuan Goodwin, “Ten telematics systems connect your car to the cloud,”
CNET, June 3, 2013, available at http://www.cnet.com/pictures/ten-telematics-systemsconnect-your-car-to-the-cloud/, site visited July 14, 2014.
33
55.
Yet, such pricing has been opposed by some in the case of Internet access platforms.
As I will now demonstrate, two-sided pricing can be an innovative and valuable component of
edge providers’ competitive strategies and can benefit consumers. The Commission should
not intentionally or unintentionally create barriers to such pricing by leaving it under a cloud
of regulatory uncertainty. Instead, the Commission should declare two-sided pricing to be
presumptively reasonable. Moreover, this presumption should be rebuttable only if the
challenging party can prove concrete harm to competition from the specific practice being
challenged.
56.
The fundamental rationale for a presumption of reasonableness is that an edge
provider’s decision to pay all or part of end users’ costs of gaining access to its services
improves the value of its services and does not degrade or reduce the value of the services
offered by rival edge providers. In other words, it is competition on the merits. Two
examples illustrate this point.
57.
First, consider an Internet access provider that employs usage-based billing in the form
of a flat fee up to a set number of gigabytes per month and incremental fees for usage above
the baseline allotment. Suppose that an edge provider pays the Internet access provider so
that household subscribers can utilize the edge providers’ service without the traffic’s
counting against the subscribers’ usage allotment. All else equal, this fact makes the edge
provider’s service more attractive to consumers. Consumers will benefit both directly (by
taking advantage of the free access) and indirectly (as competitive pressures drive the edge
provider’s rivals to improve their offerings in response). It is important to recognize that this
practice does not reduce the benefits associated with rival edge providers’ services. Indeed,
34
by creating headroom within the customer’s usage plan, one edge provider’s paying for usage
could benefit other edge providers.
58.
Second, suppose that an Internet access provider charges a set amount per unit of
traffic. If one edge provider chooses to pay the traffic-sensitive charges that would otherwise
be levied on end users utilizing its service, that conduct has no effect on what it costs end
users to utilize rival edge providers’ services. Thus, the only “harm” to rival edge providers is
that associated with increased competitive pressure.
59.
Two-sided pricing arrangements could take many forms and could potentially generate
significant consumer benefits, as the following hypothetical examples illustrates. Suppose
that a provider announced a policy of excluding the data transmitted by an antivirus app when
calculating subscribers’ usage levels. Such a strategy could be efficient given that use of
antivirus software by one end user confers benefits on other end users, similar to the way that
vaccination by one person benefits the rest of the community. Such a strategy could also
place greater competitive pressures on rival mobile wireless services providers. Despite these
benefits, a policy that forbids differentiated treatment of traffic with respect to pricing, or that
did not allow anyone other than the end user to pay for the traffic, would block this beneficial
activity. Similarly, there would be clear consumer benefits if an edge provider chose to pay
so that its subscribers would get the equivalent of a “speed boost” for its traffic. For example,
if a customer subscribes to a 25 Mbps service but an edge provider wants to pay to boost that
speed to 50 Mbps when the customer is accessing its service (say for the duration of a large
file download), the edge provider’s conduct benefits the customer and (presumably) to the
edge provider, while increasing competitive pressures on rival edge providers.
35
60.
A recent initiative by T-Mobile also illustrates how innovative pricing arrangements
can benefit consumers and competition. T-Mobile has exempted certain music services from
counting against its subscribers’ data caps. 44 Presumably, those services that are not part of
this policy will cry foul. Yet I am unaware of any reason to believe that T-Mobile has the
incentive or ability to significantly harm competition among Internet music services. Instead,
this is part of T-Mobile’s innovative “Un-carrier” strategy, which has been widely praised by
competition authorities and is disrupting the industry and promoting competition among
mobile wireless service providers.
61.
Lastly, it should again be remembered that edge providers’ interests are not fully
aligned with end users’ interests. For example, end users who do not use bandwidth-hogging
applications are harmed when forced to help pay for the consumption of those who do. End
users’ interests could be better served if edge providers faced price signals that would then
influence the design of their applications (e.g., create incentives to design their applications in
ways that make efficient use of bandwidth and reduce the burdens placed on access
networks).
44
T-Mobile’s unlimited streaming applies to several music streaming services, including
Pandora, Rhapsody, iHeartRadio, iTunes Radio, Slacker, and Spotify, and also to music
streaming services from T-Mobile partners, such as Samsung’s Milk Music and Beatport/SFX.
(T-Mobile, “T-Mobile Sets Your Music Free,” available at http://newsroom.tmobile.com/news/t-mobile-sets-your-music-free.htm, site visited July 9, 2014.)
36
VI.
IMPOSING NEW RULES ON MOBILE WIRELESS BROADBAND
INTERNET ACCESS PROVIDERS WOULD RISK DISTORTING
COMPETITION AND HARMING CONSUMERS.
62.
As discussed above, as long as it is properly implemented, continued application of the
transparency rule can promote competition and consumer welfare. This conclusion holds for
mobile wireless—as well as fixed-line—Internet access. However, there are at least three
broad reasons why application of any new rules to wireless broadband Internet access
providers poses substantial risks of distorting competition and harming consumers: 45
x
Competitive market forces already promote consumer welfare in the mobile
broadband sector. Today, several different competitors offer a variety of wireless
data and Internet access services. Competition provides incentives for service
providers to adopt network management practices and business models that promote
consumer welfare. The competitive structure of the wireless marketplace thus reduces
the potential benefits of regulation relative to the free-market outcome. Given the
strength of competition, there is relatively little reason to believe that pervasive
regulatory intervention will improve market performance.
x
Public policies that limit wireless business model innovation and network
management are especially harmful in the mobile broadband sector. Although
network management and revenue-and-service model innovation are important to the
successful operation of any communications network, they are especially important for
45
As the NPRM observes, the Commission reached a similar conclusion about characteristics of
mobile Internet access that warrant restraint in the application of new rules to mobile
broadband providers. (NPRM, ¶ 140, citing Open Internet Order, ¶¶ 93-98.)
37
wireless networks given the capacity constraints they face due to the necessity of
sharing spectrum resources and the rapidly changing nature of the industry. New
pricing models can help drive edge providers and end users to put that limited capacity
to its most high-value uses. Moreover, networks offering mobile services face
particularly complex operational issues, such as the interaction of the network with
millions of mobile transceivers that must be monitored from multiple locations and
handed off between different access points. Further, the nature of mobile access
devices may make network management practices regarding security and protection
from viruses, worms, and other malware particularly valuable. Consumers will benefit
if wireless network operators are allowed to undertake network management and to
engage in constant experimentation and innovation with respect to various aspects of
their business models without public policy creating undue obstacles.
x
Public policies that attenuate broadband mobile wireless network investment
incentives will be especially harmful. Innovation in terms of access devices, edge
services, and mobile wireless access networks has the potential to bring tremendous
new benefits to end users. Meeting the demands created by that innovation requires
very large investments by network owners. The popularity of smart phones, tablets,
and other mobile access devices has grown dramatically. 46 Traffic from wireless and
46
The Commission reports that by 2012, 55 percent of wireless consumers used a smartphone.
(“Federal Communications Commission 16th Annual Mobile Wireless Competition Report,”
March 21, 2013, at 22.) By the end of 2013, smartphones accounted for nearly 70 percent of
all mobile phone sales in the U.S. (Bank of America Merrill Lynch, “4Q13 US Wireless
Matrix,” March 26, 2014, Chart 3.) A survey by Morgan Stanley found that more than 50
38
mobile devices are expected to soon exceed traffic from wired devices, and, while PCoriginated traffic is expected to grow at an annual growth rate of 14 percent, IP traffic
from tablets and mobile phones is expected to grow by 104 percent and 79 percent,
respectively. 47 Wireless broadband providers have invested heavily in high-speed
broadband infrastructure in recent years, including the deployment of 4G technology,
improvements to existing technologies, and increased geographic coverage. A recent
White House report stated that “ … during President Obama’s first term, the annual
investment in U.S. wireless networks alone grew more than 40 percent from $21
billion to $30 billion. Projections for 2013 estimate an annual wireless network
investment at $35 billion.” 48 And significant investment in wireless infrastructure is
expected to continue. 49 Much of this investment will be in innovative network
technologies, such as 5G, that promise to deliver dramatically higher speeds to
individual users. Allowing flexibility and encouraging this innovation and investment
will make these services more useful and also increase the degree to which mobile
wireless Internet access services provide an effective competitive alternative to fixedline services. The imposition of new rules could attenuate investment incentives,
percent of respondents had a tablet in 2013, up from 20 percent the year before. (Morgan
Stanley, “4th Annual Streaming Survey,” March 13, 2013, at 1, 39.)
47
Cisco White Paper, “Cisco Visual Networking Index: Forecast and Methodology, 2012-2017,”
2013, at 2.
48
Office of Science and Technology Policy and The National Economic Council, “Four Years of
Broadband Growth,” June 2013, at 4-5.
49
Alan Pearce, J. Richard Carlson, Michael Pagano, “Wireless Broadband Infrastructure: A
Catalyst for GDP and Job Growth 2013-2017,” Prepared for PCIA – The Wireless
Infrastructure Association – by Information Age Economics, September 2013, Table A.
39
harming competition and consumers. Whatever one thinks of the potential benefits of
network neutrality regulations, those benefits must be weighed against the resulting
loss of investment incentives. 50
63.
In summary, the technical constraints faced by mobile wireless networks increase the
costs of public policies that hinder efficient network management and attenuate network
investment and innovation incentives in this evolving industry. The need for, and complexity
of, wireless network management also makes it more likely that Commission regulation of
network management will trigger those costs. At the same time, the competitive structure of
the wireless marketplace leads to forces that drive service providers to benefit consumers in
the absence of pervasive regulation. Thus, the costs of regulating broadband mobile Internet
access are likely to be high and the benefits low. For all of these reasons, application of
additional rules to mobile wireless broadband Internet access providers is particularly
problematic.
VII. CONCLUSION
64.
The Commission should employ a pro-consumer approach to policies that address the
Internet ecosystem, including broadband Internet access providers. A pro-consumer policy
50
One study that did so found that, relative to the baseline in which they were not applied, the
openness requirements attached to the C-Block license reduced the expected profitability of
the services provided with that license by 32 percent. (George S. Ford, Thomas M. Koutsky,
and Lawrence J. Spiwak, “Using Auction Results to Forecast the Impact of Wireless
Carterfone Regulation on Wireless Networks,” Phoenix Center Policy Bulletin No. 20,
(Second Edition), May 2008.) Although it is difficult to determine the precise effects, the
experience of Auction 73 demonstrates that network neutrality regulations can dramatically
lower expected network profits.
40
approach relies primarily on competitive market forces to deliver innovation and investment
with existing antitrust and consumer-protection policies serving as a backstop to correct
situations in which the market can be shown to have failed. In other words, the Commission’s
rules should let consumers choose what they want, not mandate the choices to be offered.
Instead of imposing sweeping proscriptions of evolving practices, the Commission should
adopt a flexible approach that couples case-by-case analysis with well-defined safe harbors
and presumptions of reasonableness. To do otherwise would harm consumers by distorting
competition and attenuating innovation and investment incentives.
I declare, under penalty of perjury, that the foregoing is true and correct.
Michael L. Katz
July 15, 2014
41
VIII. APPENDIX: QUALIFICATIONS
65.
I hold the Sarin Chair in Strategy and Leadership at the University of California at
Berkeley. I hold a joint appointment in the Haas School of Business Administration and in
the Department of Economics. I have also served on the faculty of the Department of
Economics at Princeton University and the Stern School of Business at New York University.
I received my A.B. from Harvard University summa cum laude and my doctorate from
Oxford University. Both degrees are in Economics.
66.
I specialize in the economics of industrial organization, which includes the study of
antitrust and regulatory policies. I regularly teach courses on microeconomics and business
strategy. I am the co-author of a microeconomics textbook, and I have published numerous
articles in academic journals and books. I have written academic articles on issues regarding
the economics of network industries, two-sided markets, systems markets, and antitrust
enforcement. I am a co-editor of the Journal of Economics and Management Strategy and
serve on the editorial boards of Information Economics and Policy and the Journal of
Industrial Economics.
67.
In addition to my academic experience, I have consulted on the application of
economic analysis to issues of antitrust and regulatory policy. I have served as a consultant to
both the U.S. Department of Justice and the Federal Communications Commission on issues
of antitrust and regulatory policy. I have served as an expert witness before state and federal
courts. I have also provided expert testimony before a state regulatory commission and the
U.S. Congress.
42
68.
From January 1994 through January 1996, I served as the Chief Economist of the
Federal Communications Commission. I participated in the formulation and analysis of
policies toward all industries under Commission jurisdiction. As Chief Economist, I oversaw
both qualitative and quantitative policy analyses.
69.
From September 2001 through January 2003, I served as the Deputy Assistant
Attorney General for Economic Analysis at the U.S. Department of Justice. I directed a staff
of approximately fifty economists conducting analyses of economic issues arising in both
merger and non-merger enforcement. My title as Deputy Assistant Attorney General
notwithstanding, I am not an attorney.
43
EXHIBIT 2
Competition in Broadband and “Internet Openness”
Declaration of Andres V. Lerner
July 15, 2015
Competition in Broadband and “Internet Openness”
Declaration of Andres V. Lerner
July 15, 2014
TABLE OF CONTENTS
I.
Introduction and Overview ...............................................................................................1
II.
Competition between Wireline Broadband Providers....................................................9
A. Most consumers have access to two or more wireline broadband providers .................9
B. Firm behavior is consistent with significant competition ............................................15
C. Consumer behavior is consistent with significant competition ...................................20
D. Consumer benefits from broadband competition.........................................................22
E. The increasing role of interconnection arrangements and content delivery
networks (CDNs) .........................................................................................................25
III.
Competition between Wireless Broadband Providers ..................................................27
A. Consumers have many options of mobile broadband providers ..................................27
B. Firm behavior is consistent with significant competition ............................................28
C. Consumer behavior is consistent with significant competition ...................................32
D. Consumer benefits from wireless broadband competition...........................................34
E. Increasing substitution between wireline and wireless broadband ..............................38
IV.
The Future Outlook for Broadband ...............................................................................40
A. Continued growth of consumer demand for broadband ..............................................41
B. Continued investment and innovation competition .....................................................46
V.
Implications of Market Competition for Proposed Regulation ...................................48
A. Limited risk that differential service models, network management, and pricing
arrangements would harm consumers and competition ...............................................49
B. Differential network management and pricing arrangements have the potential to
enhance competition and benefit consumers ...............................................................50
VI.
Conclusions .......................................................................................................................52
Appendix A: Qualifications........................................................................................................54
I.
Introduction and Overview
1.
The Federal Communications Commission (“Commission”) seeks comment on its
proposed rules for “protecting and promoting Internet openness.”1 I have been asked by counsel
for Verizon to conduct an economic analysis of competition in the broadband industry, and to
comment on the Commission’s proposed rules in light of the nature and degree of such
competition.
2.
I conclude that, given the significant competition among broadband providers, and the
nascent and rapidly changing marketplace, the rules adopted by the Commission should give
broadband providers broad flexibility to explore new service offerings and business models,
including differentiated arrangements with content providers that offer additional choices to
consumers, rather than locking in place a one-size-fits-all approach. In the absence of
demonstrable harm to competition, imposing ex ante regulation that restricts the ability of
broadband providers to experiment and deploy such business models would distort market
outcomes, reduce investment incentives, and thereby harm consumers. This is particularly true
given the rapidly changing and dynamic nature of the industry, and the significant uncertainty
concerning the services and business models that will best serve consumers going forward.
3.
Competition in the provision of broadband services. There is significant competition
among providers of broadband services. Broadband providers have made massive investments
in deploying both wireline and wireless broadband networks, and the investments necessary to
meet the growing consumer demand for broadband services are expected to continue. Both the
wireline and wireless businesses have developed, and are continuing to develop, in a competitive
manner without rigid regulations.
4.
Competition between broadband providers creates incentives for providers to implement
business models and practices that benefit consumers. And, the greater the degree of
competition, the less likely that individual broadband providers would have the incentive or
ability to engage in practices with regard to content providers that harm consumers, as
1
Federal Communications Commission Notice of Proposed Rulemaking, In the Matter of Protecting and Promoting
the Open Internet, May 15, 2014 at 4.
1|Page
consumers can choose broadband networks that best suit their purposes in terms of price and
quality attributes (including the speed and reliability of delivery of their desired content).2
5.
Wireline broadband services. Most consumers have access to two or more wireline
providers that offer broadband speeds of at least 3 Mbps (which the Commission generally
considers to be “broadband” speeds), including from cable operators, telephone companies
(DSL), and fiber providers such as Verizon and Google. In areas where wireline broadband
services are available, 99 percent of U.S. households are in census tracts with access to at least
two wireline providers that offer speeds of more than 3 Mbps. A large and growing share of the
U.S. population also has access to broadband providers offering even higher speeds. In areas
where wireline broadband services are available, 92 percent of U.S. households are in census
tracts with access to two or more wireline broadband providers offering download speeds of at
least 10 Mbps.
6.
Many areas of the U.S. are served by both next-generation, high-speed cable modem
services and all-fiber networks, like Verizon FiOS. Verizon’s all-fiber network will soon pass
approximately 70 percent of the premises in its wireline footprint. In almost all areas where
Verizon has deployed or is deploying its all-fiber network, it faces direct competition from cable
operators. Thus, within Verizon’s wireline footprint, close to 70 percent of homes soon will
have access to FiOS and high-speed cable services. Where all-fiber networks are available, there
is significant competitive rivalry between broadband providers in terms of price and quality
attributes, and consumers have access to competitive broadband services offering speeds of
hundreds of megabits per second.
7.
Other fiber-based deployments—such as “fiber-to-the-neighborhood” (“FTTN”, also
referred to as “fiber-to-the-node”)—also are increasingly available and offer higher speeds than
traditional DSL services. And, in areas where fiber is not yet available, most consumers also
have access to DSL as well as cable. While the speeds and capabilities of DSL services are more
limited than those of fiber or cable networks, DSL does offer a less expensive option for
consumers who are more price-sensitive or do not require particularly high-speed broadband, and
therefore can exert competitive discipline on cable providers.
2
I approach my analysis and conclusions from the perspective of consumer welfare, and the viewpoint that
promoting competition is fundamental to enhancing consumer welfare.
2|Page
8.
Competition has prompted providers to continue to make massive investments in
deploying and upgrading next-generation broadband technology that offers consumers faster and
faster broadband speeds. The introduction of fiber-based broadband services over fiber-to-thehome (or “FTTH”) networks starting around a decade ago ignited the race to make nextgeneration broadband services available to consumers. FTTH networks, while expensive to
deploy, offer virtually unlimited capacity to meet consumer demand for higher speeds and lower
latency. Verizon was the first to deploy FTTH on a wide scale, starting in 2004, investing
around $23 billion in its FiOS network. FiOS is the largest provider of FTTH in the United
States, passing nearly 19 million households as of early 2014. Google has more recently begun
to deploy FTTH in select cities, and is considering the possibility of expanding in many others.
Other providers, such as AT&T’s U-Verse, have extended fiber closer to the home in order to
achieve higher speeds than traditional DSL services (but not as high as FTTH).
9.
Cable operators responded to the deployment of fiber networks by also offering very
high-speed broadband services, including by rolling out DOCSIS 3.0 technology. Because this
deployment only required upgrades to the cable operators’ existing plant, as compared to the
relatively higher expense of constructing new fiber facilities, these services quickly became more
widely available than fiber services. These next-generation cable modem services currently
reach over 85 percent of the U.S. population. While next-generation cable services have greater
capacity limitations than fiber networks, particularly for upstream speeds, these services offer
high download speeds exceeding the demands of most of today’s consumers, with services
currently available that offer download speeds of greater than 100 Mbps.
10.
The competitive rivalry between wireline broadband access providers has led to falling
(quality-adjusted) prices, higher broadband speeds, increased availability of high-speed
broadband, greater consumer choice of products and packages, and other improvements in
quality. There is no reason to believe that this competitive rivalry will diminish.
11.
Wireless broadband services. There is vigorous competition among wireless providers.
In this still nascent sector, wireless providers have made significant investments to deploy highspeed broadband services to consumers, including “fourth-generation” (“4G”) LTE technology,
and to improve network coverage and capacity. Today, about 95 percent of consumers have
access to broadband download speeds of greater than 10 Mbps over wireless networks.
3|Page
Consumers can obtain wireless broadband from many providers, with almost 90 percent of the
U.S. population having access to four or more wireless broadband carriers.
12.
Wireless providers compete fiercely for customers on the basis of price, as well as with
respect to important aspects of the wireless ecosystem, including the provision of valuable
services, handset devices, operating systems, applications, and content. This competition among
wireless providers has led to important consumer benefits. Mobile wireless speeds continue to
rise, prices per megabyte of data continue to fall, and consumers have access to a wide and
increasing variety of devices and plans. Even as the rollout of 4G LTE continues, the industry is
already looking towards the next round of innovation and investment to further expand the
capabilities of wireless broadband networks—innovations that could make wireless broadband
more competitive with higher speed wireline services for more customers.
13.
The changing competitive landscape. Industry developments in the past few years have
transformed the competitive landscape. As a result of these industry developments, there are
more substitutes for every part of the broadband network than traditional definitions would
suggest.
14.
The last mile. With regard to the last mile, the lines between traditional broadband
network distinctions—wireline and wireless—have been blurring, and are expected to continue
to blur. Consumers increasingly connect to wireline and wireless networks for many of the same
uses, and increasingly substitute between wired and wireless devices and networks. These
changes in broadband usage have been spurred by significant investments and other industry
developments. First, significant improvements in speed and capacity of wireless networks have
allowed consumers to perform many of the same tasks on wireless devices as they perform on
computers connected via wireline networks. Although generally slower than wireline broadband
options such as fiber and DOCSIS 3.0, 4G LTE network providers typically offer average
download speeds of 5 to 12 Mbps that are sufficient for many uses, including streaming video,
with the added benefit of mobility. While mobile broadband services today are unlikely to offer
the same speeds and capabilities as next-generation wireline options like cable and fiber, wireless
broadband is a rapidly-developing sector and wireless speeds have the potential to increase
further with advancements in technology and the deployment of additional spectrum, both of
4|Page
which require additional investments. Second, the advent of advanced wireless devices,
particularly tablets, has blurred the distinction between wireless and wireline devices.
15.
As a result, wireless broadband services already are, to some degree, a competitive
alternative to wireline networks for some consumers, further increasing competitive pressures to
meet consumer demands, and diminishing the incentive and ability of wireline broadband
providers to adopt business practices disfavored by consumers. Wireless broadband has the
potential to become an alternative to wireline broadband for more and more consumers, and for
more uses, in the future if the rapid pace of technology improvements and innovation continue.
However, wireless broadband is still a nascent industry that will require significant further
investments, development, and innovation in order for this to happen. Accordingly, providing
flexibility for wireless broadband providers to experiment with and implement different service
offerings and business models is particularly important to enhancing the competitiveness of both
the wireline and wireless industries.
16.
The Internet backbone. With respect to the Internet backbone, content providers are
using various ways to enable faster and more reliable content delivery. Networks and content
providers are connecting directly more often, and in more places. The exchange of traffic over
these connections is governed by an increasing variety of arrangements, such as “paid peering”
where one party compensates another for connection and access to its network, and “quality of
service” agreements that guarantee a certain level of transit quality by the network provider (e.g.,
low latency or packet loss).
17.
The exchange of traffic is also facilitated by firms other than broadband networks. For
instance, content delivery networks (“CDNs”) sell services to content and application providers
that improve the speed and quality at which their content can be accessed by users, generally by
maintaining a large number of geographically-dispersed servers containing cached content that
connect to the Internet near the location of customers. CDNs reduce the number of potential
congestion points for delivering content from the providers’ data centers to end users. These
ways of bypassing the backbone of the Internet and traditional forms of content delivery are
becoming common. Many content providers use third party CDNs such as Akamai, Limelight,
and Chinacache. Moreover, very large content providers such as Netflix, Google, and Amazon
have developed their own proprietary CDNs.
5|Page
18.
To the extent that CDNs and various interconnection arrangements are substitutes to
network management or differentiated arrangements for delivery over the last mile by broadband
providers from the perspective of content providers (e.g., for purposes of improving speed,
latency, and reliability), these CDNs also may constrain the business practices of broadband
providers. That is, even though CDNs manage traffic on different parts of the network, they may
provide content providers an alternative means of increasing the speed, reducing latency, and/or
improving reliability by which their content is delivered to users. This may constrain further the
ability of broadband providers to engage in practices related to their last-mile facilities that could
harm competition or consumers.
19.
The development of CDNs and other innovative network arrangements also highlights
the ability of the industry participants to adapt to the changing competitive landscape in creative
and procompetitive ways, and to keep innovating in ways that create important consumer
benefits. These developments have been achieved by relying on competitive market forces
rather than through proscriptive or prescriptive regulation.
20.
The future outlook for broadband. Despite the staggering investments by wireline and
wireless broadband providers and the competitive successes of the broadband industry to date,
there are significant opportunities and challenges ahead as broadband providers compete to bring
innovative products and services to consumers. The rapid growth in demand for broadband of
the past few years is expected to continue, if not accelerate. This growth in demand for
broadband will largely be driven by increasing demand for certain content, particularly video,
which is predicted to account for almost 70 percent of all Internet traffic in the next few years.
21.
Similar trends are occurring in the wireless area, and these trends will create even greater
challenges for wireless providers due to the significant capacity constraints they face, including
as a result of the inherently limited amount of spectrum available. The launch of the iPhone in
June 2007 and subsequent developments have led to an explosion in mobile broadband usage, as
mobile Internet-connected devices proliferated. The rapid growth in consumer demand for
wireless broadband is expected to continue, with more and more of this growth generated by
specific types of applications such as video. In the next few years, the volume of mobile video
traffic is expected to increase tenfold, and more than half of all mobile data traffic is projected to
come from video usage.
6|Page
22.
Wireline and wireless broadband providers are expected to make significant
infrastructure investments in the next few years to meet this growth in demand. But the biggest
challenge is that the future of broadband use is highly uncertain. What content and applications
will be offered, what consumers will demand, and the business models that will develop to meet
consumer demands are largely unknown. The relatively short history of the Internet is filled with
examples of firms introducing new business models that have changed how consumers use the
Internet, and examples of leading companies that have gone by the wayside as technology and
consumer tastes changed. The tremendous innovation that has occurred in the past few years is
expected to continue, and future innovation may dwarf the advances of the past. This innovation
will require experimentation with new services, business models, pricing structures, and network
management practices.
23.
Implications for the Commission’s proposed rules. Given the level of competition in the
broadband industry, and the nascent and rapidly changing broadband environment, the
Commission should give broadband providers wide flexibility to experiment with novel service
offerings and business models, including differential pricing and network management
arrangements. Both the wireline and wireless industries have developed, and are continuing to
develop, in a competitive manner. The current competitive environment, and massive historical
and planned investments in deploying broadband networks to meet consumer demands, have
been achieved by relying on competitive market forces, not through rigid regulation. Without
heavy-handed regulation, the marketplace has shown considerable ability to meet the challenges
brought on by changes in consumer demand and in the services offered by edge providers over
broadband networks. There is no reason to believe that a considerable change in the regulatory
approach is now necessary to ensure continued competitive rivalry, innovation, and investments.
24.
Given the competitiveness of the wireline and wireless broadband industries, there is very
limited risk that providers would adopt pricing and network management approaches that would
harm competition and consumers. There is no evidence that broadband providers face
insufficient competitive constraints, which is a necessary condition for firms to have the
incentive and ability to enter into vertical business arrangements that harm competition and
consumers. The risk of losing subscribers from inefficient practices is likely to provide a
significant competitive constraint on broadband providers. And, if anything, competition
between broadband providers is likely to intensify in the coming years.
7|Page
25.
While the risks of competitively harmful business practices are very limited, differential
network management and pricing arrangements have the potential to generate significant benefits
for competition and consumers. Two-sided pricing approaches provide one example of the types
of arrangements that could generate important consumer benefits. Most directly, positive prices
to content providers would tend to lower prices to consumers. The economics literature on “twosided markets” shows that there is no economic basis that mandating “free” access to one side of
the market (in the current context, content providers) is good for competition, consumer welfare,
or innovation. Although, to date, the competitive equilibrium seems generally to be one where
last-mile broadband providers have not directly entered into pricing arrangements with content
and application providers (such as prioritization agreements), industry developments may make
two-sided pricing arrangements beneficial.
26.
As a result, there is little benefit and high risk of implementing regulation prohibiting
certain pricing and network management strategies before demonstrating harm to competition
and consumers that could arise from such practices. Restrictive ex-ante regulation would be
especially inefficient because the broadband industry is rapidly changing and highly dynamic,
and there is significant uncertainty regarding the future of the broadband industry. It is not
possible to predict consumer demand, or the types of business models that undoubtedly will
emerge. As a result, a one-size-fits-all approach is likely to deny consumers choices that they
would otherwise find attractive.
27.
In sum, there is no reason to believe that ex ante regulation that bans or restricts specific
business models adopted by broadband providers would enhance competition and consumer
welfare. In fact, imposing regulation that restricts the ability of providers to experiment with or
adopt legitimate business arrangements would impose considerable costs, distort competition and
market outcomes, and reduce investment incentives, all to the detriment of consumers. Imposing
ex ante regulation on the wireless industry would pose an even greater threat to competition and
consumer welfare. Rather than trying to impose a particular structure on service offerings or
business models implemented by broadband providers, the Commission should adopt policies
that give providers flexibility to experiment with and implement a wide range of approaches.
8|Page
II.
Competition between Wireline Broadband Providers
A. Most consumers have access to two or more wireline broadband providers
28.
Most consumers have access to two or more wireline providers offering broadband with
speeds of at least 3 Mbps, which the Commission generally considers to be “broadband” speeds.3
These broadband networks include the coaxial networks of cable operators, DSL offered by
many telephone companies, and fiber networks offered by firms such as Verizon and Google.
Data provided by the Commission shows that, in areas where wireline broadband services are
available, 99 percent of U.S. households are in census tracts that have access to at least two
wireline providers that offer speeds of more than 3 Mbps, compared to 74 percent in 2010, as
shown in Figure 1 below.4
3
The Commission generally considers “broadband” download speeds to be those of at least 3 to 4 Mbps. (See, e.g.,
Federal Communications Commission Industry Analysis and Technology Division, Wireline Competition Bureau ,
“Internet Access Services: Status as of December 31, 2012,” December 2013 at 4.) The Commission also has noted
that 1.5 Mbps is a sufficient download speed for most consumers’ online activities, including email, playing video
games, and streaming music, videos, or movies. The Commission notes that high-definition (HD) video streaming
requires a minimum download speed of 4 Mbps. (Federal Communications Commission Broadband Speed Guide,
available at http://www.fcc.gov/guides/broadband-speed-guide.)
4
Figures 1 and 2 are based on areas with at least one wireline broadband provider, which cover about 96.7 percent
of the U.S. population. (Federal Communications Commission National Broadband Map,
http://www.broadbandmap.gov/summarize/nationwide. Data as of June 30, 2013.) The data identifies providers
operating anywhere in a given census tract, so it is possible that not every home in the census tract is passed by
every provider operating in the area. Overall, 88 percent of the U.S. population has access to two or more wireline
broadband providers, and 56 percent has access to three or more wireline broadband providers. (Federal
Communications Commission National Broadband Map, http://www.broadbandmap.gov/summarize/nationwide.
Data as of June 30, 2013.)
9|Page
Figure 1: Share of U.S. Households with Wireline Broadband of at Least 3 Mbps
100%
97%
90%
Share of Total U.S. Households
80%
70%
99%
84%
74%
60%
1 Provider
50%
2+ Providers
40%
30%
23%
20%
14%
10%
3%
1%
0%
2010
2011
2012
2013
Notes: Shares reflect households in census tracts where providers report offering wireline broadband service at speeds of 3Mbps/786Kbps (download/upload) or
higher. Data as of June of each year.
Sources: FCC Wireline Competition Bureau, Industry Analysis and Technology Division, “Internet Access Services” Reports, 2010 (March 2011), Figure 3(a);
2011 (June 2012), Figure 3(a); 2012 (May 2013), Figure 5(a); 2013 (June 2014), Figure 5(a).
29.
A large share of the U.S. population has access to multiple broadband providers offering
much higher speeds. In 2013, 92 percent of U.S. households in areas with some form of wireline
broadband were in census tracts where two or more wireline broadband providers offered at least
10 Mbps download speeds, as shown in Figure 2 below, a considerable increase from 23 percent
of households in 2010.5 As I discuss below, this rapid increase in the availability of high-speed
broadband in the past few years was driven in large part by investments in deploying fiber
networks and improving cable network technologies.
5
Federal Communications Commission Industry Analysis and Technology Division, Wireline Competition Bureau,
“Internet Access Services: Status as of June 30, 2013,” June 2014, Figure 5(a).
10 | P a g e
Figure 2: Share of U.S. Households with Wireline Broadband of at Least 10 Mbps
100%
92%
90%
Share of Total U.S. Households
80%
70%
60%
60%
55%
51%
1 Provider
42%
2+ Providers
50%
40%
31%
30%
23%
20%
7%
10%
0%
2010
2011
2012
2013
Note: Shares reflect households in census tracts where providers report offering wireline broadband service at speeds of 10Mbps/1.5Mbps (download/upload) or
higher. Data as of June of each year.
Sources: FCC Wireline Competition Bureau, Industry Analysis and Technology Division, “Internet Access Services” Reports, 2010 (March 2011), Figure 3(a);
2011 (June 2012), Figure 3(a); 2012 (May 2013), Figure 5(a); 2013 (June 2014), Figure 5(a).
30.
Coaxial networks: In most areas, cable providers were the initial providers of high-speed
broadband, using networks of coaxial cable.6 Today, cable broadband Internet access is
available to 88 percent of U.S. consumers.7 Cable broadband usage in the U.S. continues to
expand, with subscriptions growing to more than 51 million, accounting for over 53 percent of
total broadband subscribers.8
31.
DSL networks: Local telephone companies have traditionally offered consumers
broadband access over their copper phone wire networks, via a technology called digital
6
Most cable operators use hybrid fiber-coaxial (“HFC”) networks, where fiber is used for transmission from major
distribution stations to smaller neighborhood “nodes,” and then coaxial cable runs from the nodes to subscribers’
homes. (See, e.g., Timothy Brophy, Steven Condra, Martin Mattingly, Ron Hranac, & Leonard Ray, “FTTH
Evolution of HFC Plants,” Cisco, September 2011 at 3.)
7
Federal Communications Commission National Broadband Map, available at
http://www.broadbandmap.gov/summarize/nationwide. Data as of June 30, 2013.
8
Strategy Analytics, “Service Provider Strategies (SPS), North America Broadband Forecast,” 2014 at 3.
11 | P a g e
subscriber line (“DSL”). As of 2013, DSL was available to 89 percent of U.S. consumers.9
Some telephone companies have improved their DSL networks by deploying fiber in some parts
of the network. For instance, AT&T’s U-verse uses fiber-to-the-neighborhood (“FTTN”, also
referred to as “fiber-to-the-node”), and delivers broadband into homes over traditional copper
wire DSL technology.10 The copper wiring to the home limits the speeds achievable by U-Verse
and other FTTN networks, although extending fiber farther into the field increases speeds
compared to traditional DSL services.
32.
Fiber networks: Within the past decade, telecommunications firms such as Verizon have
begun offering very high-speed broadband by deploying fiber-to-the-home (“FTTH”) broadband
networks. Verizon FiOS, launched in 2004, was the first major FTTH undertaking in the U.S. It
is reported that Verizon has invested roughly $23 billion in building out the FiOS network.11 As
of March 2014, Verizon’s all-fiber network passed 18.9 million premises (and was available to
over 15.5 million of those), with around 6.2 million broadband customers.12 Verizon plans to
have FiOS available to 70 percent of households in its wireline footprint.13 Verizon’s FiOS
network currently offers broadband download speeds as high as 500 Mbps (up from 50 Mbps in
2007).14 By continuing expansion of its fiber network, and meeting consumer demands for highspeed broadband access, Verizon’s FiOS has steadily increased its penetration since launch (see
Figure 3 below).15
9
Federal Communications Commission National Broadband Map, available at
http://www.broadbandmap.gov/summarize/nationwide. Data as of June 30, 2013.
10
“How AT&T U-verse TV is Delivered,” ATT.com, available at
https://www.att.com/Common/about_us/files/pdf/HowUverseIsDelivered_2-22.pdf.
11
See, e.g., Malia Spencer, “Verizon plans more hires as it finalizes new FiOS tech center,” Pittsburgh Business
Times, June 10, 2011, available at http://www.bizjournals.com/pittsburgh/print-edition/2011/06/10/verizon-morehires-new-fios-tech-center.html?page=all.
12
“Verizon Communications Investor Quarterly 1Q 2014,” April 24, 2014 at 7; Verizon Communications 10-Q,
April 29, 2014 at 31.
13
Thomson Reuters StreetEvents, “VZ - Verizon at Jefferies Global Technology, Media & Telecom Conference,”
May 6, 2014 at 14, transcript available at
http://www.verizon.com/investor/DocServlet?doc=jefferies_conf_vz_trans_2014.pdf.
14
See Figure 5 infra.
15
An analyst reported that two thirds of new FiOS subscribers were either new broadband subscribers, or previous
cable subscribers. (Andrew Schmitt, “Strong Competitive Impact From Verizon's FiOS Broadband,” Seeking
Alpha, May 4, 2006, available at http://seekingalpha.com/article/10074-strong-competitive-impact-from-verizonsfios-broadband-vz.)
12 | P a g e
Figure 3: Verizon FiOS Penetration Rates
Penetration Rate (FiOS Connections As Share of Homes Passed)
45%
40%
35%
30%
25%
20%
15%
10%
5%
0%
2006
2007
2008
2009
2010
2011
2012
2013
Sources: Verizon Communications 2006 10-K at 47; Verizon Communications 2007 10-K at 39; Verizon Communications 2008 10-K at 43; Verizon
Communications 2009 10-K at 70; Verizon Communications 2010 10-K at 97; Verizon Communications 2011 10-K at 53; Verizon Communications 2012 10-K
at 53; Verizon Communications 2013 10-K at 61.
33.
More recently, other firms also have begun to deploy FTTH networks. For instance,
Google Fiber offers gigabit speeds through its FTTH network. A recent industry report
concluded that “Google’s aim is to build a competitive, profitable, fixed network covering a
significant portion of US households.”16 Although Google Fiber is currently available only in
Kansas City, Missouri, and Provo, Utah, Google is also building out its Google Fiber network in
Austin, Texas and discussing the possibility of further expansion with 34 other U.S. cities and
towns.17 In areas where it has been deployed, Google Fiber has successfully competed against
cable providers.18 Other firms that have deployed FTTH networks in competition with cable
16
Jeff Baumgartner, “Google Fiber Capturing 75% Of Homes Passed In KC: Study,” Multichannel News, May 6,
2014, available at http://www.multichannel.com/news/technology/google-fiber-capturing-75-homes-passed-kcsurvey/374365, discussing Carlos Kirjner and Peter Paskhaver, “Google Fiber: Scale Matters – How Large Could it
Be? How Fast Could It Grow? Introducing Bernstein’s BIGR Model,” Bernstein Research, May 7, 2014.
17
Google Fiber expansion plans available at https://fiber.google.com/newcities/.
18
For instance, analysts estimate the penetration of Google Fiber in Kansas City and surrounding areas could
increase to 65 percent in three to five years. (Alistair Barr, “Google Fiber Gets Snapped Up in Kansas City,” Wall
Street Journal Digits, May 7, 2014, available at http://blogs.wsj.com/digits/2014/05/07/google-fiber-gets-snappedup-in-kansas-city/.)
13 | P a g e
providers include HBC, PAXIO, and Smithville. There are also many smaller telephone
companies that offer FTTH,19 and AT&T has plans to invest in FTTH in several metropolitan
areas.20 Currently, FTTH is available to about 12 percent of U.S. consumers.21
34.
Overall, fiber broadband usage has continued to show significant growth. FTTH and
FTTN are expected to increase from 17 million subscribers in 2013 to 27 million subscribers in
2019.22 By 2019, almost 25 percent of broadband subscribers are expected to have service from
a fiber-based network, as shown in Figure 4 below. This increase in fiber subscribers includes
gains vis-à-vis cable providers, as well as projected upgrades for consumers with traditional DSL
networks to FTTN and FTTH.
19
Telecommunications Industry Association, “TIA’s 2014-2017 ICT Market Review and Forecast,” 2014 at 3-27.
20
Marguerite Reardon, “AT&T to Take Gigabit Broadband to 21 New Metro Areas,” CNET, April 21, 2014,
available at http://www.cnet.com/news/at-t-to-take-gigabit-broadband-to-21-new-metro-areas/.
21
Federal Communications Commission National Broadband Map, available at
http://www.broadbandmap.gov/about-provider/verizon-communications-inc./nationwide/. Data as of June 30, 2013.
22
Strategy Analytics, “Service Provider Strategies (SPS), North America Broadband Forecast,” April 2014 at 11.
The data for FTT(x) include both FTTH providers (e.g., Verizon and Google Fiber) and other non-cable providers
using fiber, such as AT&T (whose U-verse network uses FTTN).
14 | P a g e
Figure 4: U.S. Broadband Subscriber Shares by Technology
70%
Projection
Share of All U.S. Broadband Subscribers
60%
50%
40%
Cable
30%
DSL
FTT(x)
Wireless Only
20%
10%
0%
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
Notes: FTT(x) includes networks that use fiber in any part of the network. Markets with established and significant LAN or Powerline service subscriptions are
included in the FTT(x) category. Wireless Only includes homes with primary access platform that is 3G/4G, fixed/mobile Wi-Max, or MMDS/LMDS satellite.
Source: Strategy Analytics (SGS), "North American Broadband Forecast 2014," April 2014, tab USA.
B. Firm behavior is consistent with significant competition
35.
Price and quality competition: There is significant competitive rivalry between wireline
broadband providers in terms of price and quality attributes. Evidence of the strong competitive
rivalry between cable and fiber can be observed from the deployment of fiber networks,
particularly Verizon FiOS, and more recently by other firms such as Google. The deployment of
fiber networks has incentivized cable operators to compete by lowering prices and improving
quality. In areas where FiOS was first launched, cable incumbents responded by lowering
prices.23 An analyst report stated that “[t]he rollout of Verizon’s FiOS service in select markets
23
For instance, when Verizon launched its FiOS fiber service in 2005 in Texas, incumbent cable provider Charter
Communications reportedly dropped its prices shortly after. (Steve Titch, “Texas Gets Cable Competition, Lower
Prices,” Heartland, May 1, 2006, available at http://news.heartland.org/newspaper-article/2006/05/01/texas-getscable-competition-lower-prices.) Later in 2008 as Verizon began to expand its FiOS system in New York, it was
reported that cable rivals responded by “cutting prices on voice services and developing new technology that can
match the data speeds of FiOS and cram in more TV channels.” (Saul Hansell, “Verizon FiOS: A Smart Bet or a
Big Mistake?” The New York Times, August 18, 2008, available at
http://www.nytimes.com/2008/08/19/technology/19fios.html?pagewanted=all&_r=0.)
15 | P a g e
has elicited thinly advertised, yet highly competitive pricing responses for incumbent cable
providers.”24 A few years later, a 2010 article noted that FiOS “entry into an area leads to lower
cable prices.”25 A more recent example of this competitive rivalry occurred in late 2013, with
Time Warner Cable cutting the price in half for its low-speed tier broadband offering and
doubling the download speeds of both its low- and high-speed tiers in order to compete with
Verizon FiOS.26
36.
Google’s recent deployment of fiber networks provides further evidence of this
competitive rivalry, with cable providers attributing their competitive responses consisting of
faster broadband speeds at lower prices to the deployment of Google Fiber. In Provo, Utah,
where Google Fiber recently launched, Comcast reportedly “slashed” the price for a bundle of
cable video and 105 Mbps broadband to $120, the same price as Google’s high-speed broadband
and Google TV package.27 Additionally, Comcast reportedly offered a 250 Mbps speed service
to compete with Google Fiber in Provo.28 In Kansas City, the entry of Google’s Fiber in 2012
compelled Time Warner Cable to increase download speeds in all of its broadband service
packages at no extra charge to consumers.29 Analysts believe Google Fiber will “encourage
current broadband network providers to get more aggressive with advanced rollouts.”30
24
Steve Titch, “Texas Gets Cable Competition, Lower Prices,” Heartland, May 1, 2006, available at
http://news.heartland.org/newspaper-article/2006/05/01/texas-gets-cable-competition-lower-prices.
25
Peter Svensson, “Verizon winds down expensive FiOS expansion,” Seattle Times, March 26, 2010, available at
http://seattletimes.com/html/businesstechnology/2011449152_apustecverizonfios.html.
26
“TW Cable Boosts Top Internet Speed,” Santa Monica Daily Press, October 28, 2013. See also, Deutsche Bank,
“A Closer Look at FiOS,” April 1, 2014; Evercore Partners, “FiOS Market Level Analysis Points to Further Pressure
for Cablevision,” September 15, 2013.
27
“Comcast has slashed the pricing on its triple-play package in Provo, Utah, in advance of Google Fiber's launch in
the city later this year.” (Steve Donohue, “Report: Google Fiber threat prompts Comcast to slash prices in Provo,”
FierceCable, August 15, 2013, available at http://www.fiercecable.com/story/report-google-fiber-threat-promptscomcast-slash-prices-provo/2013-08-15#ixzz35ZHxJD83.) See also, “Google Fiber’s Impact on US Broadband,”
Strategy Analytics, October 7, 2013.
28
Jeff Baumgartner, “Google Fiber Kicks Off Provo Installs,” Multichannel News, November 13, 2013, available at
http://multichannel.com/news/content/google-fibers-kicks-provo-installs/356977.
29
Todd Spangler, “TWC Kicks Up Broadband Download Speeds,” Multichannel News, December 20, 2012,
available at http://multichannel.com/news/cable-operators/twc-kicks-broadband-download-speeds/326108.
30
Bank of America Merrill Lynch, “Google Inc., Speed Matters, Going Big on Fiber,” February 19, 2014 at 1.
16 | P a g e
37.
Although DSL speeds are generally not comparable to speeds achieved over cable or
fiber networks, DSL does offer a less expensive option for consumers who are more pricesensitive or do not require particularly high-speed broadband.
38.
Investment and innovation competition: In dynamic, high-technology industries like the
broadband industry, primary dimensions of competition are investment and innovation, as firms
strive to create and offer better products and services to consumers. In the broadband industry,
competition has prompted providers to continue to invest massive sums of money both in
deploying next-generation broadband technology that offers consumers faster broadband speeds
and in further upgrading and maintaining those networks to handle the rapidly growing volumes
of traffic. In the past decade, the wireline broadband industry has invested hundreds of billions
of dollars to offer consumers better-quality services—from 1996 to 2012, the U.S. wireline
broadband industry invested approximately $660 billion.31
39.
Some broadband providers, particularly Verizon, have invested significant capital to
deploy extensive broadband networks that achieve very fast broadband speeds by extending fiber
to the home. It is reported that Verizon has invested roughly $23 billion in building out the FiOS
network.32 As mentioned, Verizon’s FiOS has become a strong competitor to cable operators in
many areas of the country. FiOS will soon be available in 70 percent of Verizon’s wireline
footprint.33 In almost all areas where Verizon is deploying its all-fiber network, it faces direct
competition from cable operators.
40.
Google has built out its fiber network in two U.S. cities, is currently building out its
network in another, and is considering the possibility of deploying all-fiber networks in many
others. DSL providers have also been upgrading their legacy DSL infrastructure.34 For example,
31
U.S. Telecom Research Brief, “Updated Capital Spending Data Show Rising Broadband Investment in Nation’s
Information Infrastructure,” November 4, 2013 at 2.
32
See, e.g., Malia Spencer, “Verizon plans more hires as it finalizes new FiOS tech center,” Pittsburgh Business
Times, June 10, 2011, available at http://www.bizjournals.com/pittsburgh/print-edition/2011/06/10/verizon-morehires-new-fios-tech-center.html?page=all.
33
Thomson Reuters StreetEvents, “VZ - Verizon at Jefferies Global Technology, Media & Telecom Conference,”
May 6, 2014 at 14, transcript available at
http://www.verizon.com/investor/DocServlet?doc=jefferies_conf_vz_trans_2014.pdf.
34
SNL Kagan, “Media Trends Actionable Metrics, Benchmarks & Projections For Major Media Sectors,” December
2013 at 105.
17 | P a g e
as described above, AT&T’s U-verse employs a combination of fiber (to the node) and copper
wire (to the subscriber’s home), and has announced plans to deploy its 1 Gbps high-speed
“GigaPower” all-fiber (FTTH) broadband network in a number of locations.35
41.
These investments in fiber networks have spurred competition to deploy faster and faster
broadband, and have compelled cable companies to upgrade their own networks in order to
provide high-speed broadband services.36 The cable providers responded to Verizon’s
deployment of FiOS by deploying DOCSIS 3.0 technology. As a Wall Street Journal article
reported in 2008, analysts believed DOCSIS 3.0 “will allow the cable industry to compete on a
more even footing with telecom giant Verizon Communications Inc., which is aggressively
marketing a high-performance fiber-optic network called FiOS that offers much faster Internet
connection speeds than cable modems can currently deliver.”37
42.
Cable firms have continued to roll out DOCSIS 3.0 over the past five years, and this
high-speed broadband technology is now available to more than 85 percent of households in the
U.S.38 This upgraded infrastructure will also facilitate increasing broadband speeds in the future
(so-called “DOCSIS 3.1”), with speeds closer to 1 Gbps.39 Cable companies also have begun
deploying FTTH in some areas in order to compete with FiOS.40
35
Marguerite Reardon, “AT&T to Take Gigabit Broadband to 21 New Metro Areas,” CNET, April 21, 2014,
available at http://www.cnet.com/news/at-t-to-take-gigabit-broadband-to-21-new-metro-areas/.
36
An industry analyst recently noted that “Verizon FiOS and AT&T U-Verse have already started pushing up
speeds in other areas to create more pressure on the cable operators. Cable operators are responding, or in some
cases leading, by deploying DOCSIS 3.0 solutions with 100Mbps and greater speeds.” (Strategy Analytics, “Google
Fiber’s Impact on US Broadband,” October 7, 2013.)
37
Vishesh Kumar, “Cable Prepares an Answer to FiOS,” The Wall Street Journal, February 14, 2008, available at
http://online.wsj.com/news/articles/SB120295689385867313.
38
National Cable & Telecommunications Association Industry Data, available at https://www.ncta.com/industrydata.
39
Jeff Baumgartner, “DOCSIS 3.1 Speeds Ahead,” Multichannel News, April 28, 2014, available at
http://www.multichannel.com/news/news-articles/docsis-31-speeds-ahead/374179. See also, Telecommunications
Industry Association, “TIA’s 2014-2017 ICT Market Review and Forecast,” 2014 at 3-25: “The rollout of DOCSIS
3.0 with channel bonding has increased maximum possible speeds, and the DOCSIS 3.1 specification, expected to
be adopted in 2014, will enable speeds of up to 10 Gbps downstream and 1 Gbps upstream without the need to run
FTTH.”
40
For instance, in September 2013, two months after FiOS began offering a 500 Mbps connection in the same areas,
Comcast offered an Internet plan with a top download speed of 505 Mbps using FTTH. Comcast’s fiber offering
was available in select cities in the northeast U.S., including Washington, D.C., Philadelphia, Boston, Hartford,
Baltimore and Richmond. (“Comcast Fights Fiber with Fiber,” Multichannel News, September 23, 2013.) See also,
18 | P a g e
43.
In turn, Verizon has continued to make investment in innovations to increase the speeds
of its broadband offerings. Maximum download speeds for Verizon FiOS have increased by a
factor of 10 since 2006, as shown in Figure 5 below, with most of those increases since 2011.
These enhancements were achieved through technological investments, including the launch of
FiOS “Quantum” service in 2012, which increased available maximum download speeds to 300
Mbps,41 and upgraded to 500 Mbps in 2013.42
Figure 5: Verizon FiOS Maximum Download Speeds
600
500
Maximum Download Speed (Mbps)
500
400
300
300
200
150
150
2010
2011
100
50
50
50
50
2006
2007
2008
2009
0
2012
2013
Sources: Verizon Communications 2006 Annual Report at 4; Verizon Communications 2007 Annual Report at 13; Verizon Communications 2008 10-K at 11;
Verizon Communications 2009 10-K at 13; Verizon Communications 2010 10-K at 10; Verizon Communications 2011 10-K at 8; Verizon Communications 2012 10K at 8; Verizon Communications 2013 10-K at 8.
44.
The Commission has recognized the significant role of competition in creating incentives
to invest, and the benefits to consumers that this investment has generated. In its National
“Comcast Launches ‘Extreme 505’ Internet Tier in Chicago area,” Comcast Greater Chicago Region, May 30, 2014,
available at http://comcastgcr.com/2014/05/30/comcast-launches-extreme-505-internet-tier-in-chicago-area/.
41
Emily Price, “Verizon Launches New FiOS Quantum Service, Offers Speeds Up To 300 Mbps,” Mashable, June
17, 2012, available at http://mashable.com/2012/06/17/verizon-fios-quantum/.
42
Chris Welch, “Verizon rolls out fastest FiOS tier yet with 500Mbps downloads, 100Mbps uploads,” The Verge,
July 22, 2013, available at http://www.theverge.com/2013/7/22/4546286/verizon-rolls-out-fastest-fios-quantum-tier500-100.
19 | P a g e
Broadband Plan, the Commission noted that “[d]ue in large part to private investment and
market-driven innovation, broadband in America has improved considerably in the last decade.
More Americans are online at faster speeds than ever before.”43 Similarly, a recent White House
report on broadband stated:
Responding to the increasing consumer demand for services accessed through
broadband, the private sector has been driving important advances in
infrastructure and technology. U.S. telecommunications firms have made
significant investments in infrastructure; for example, just two of the largest U.S.
telecommunications companies account for greater combined stateside investment
than the top five oil/gas companies, and nearly four times more than the big three
auto companies combined. In fact, since President Obama took office in early
2009, nearly $250 billion in private capital has been invested in U.S. wired and
wireless broadband networks. In just the last two years, more high-speed fiber
cables have been laid in the United States than in any similar period since 2000.44
C. Consumer behavior is consistent with significant competition
45.
The behavior of consumers also is consistent with the existence of significant competitive
rivalry among broadband providers, and that is particularly the case in areas where two or more
wireline networks are available that are capable of supporting high broadband speeds, such as
cable and fiber. In competitive industries, consumers generally evaluate competitive options,
and choose the best provider for their needs. When better-quality and/or lower-price products or
services become available (or when consumers become aware of the availability of such products
and services), consumers in competitive markets are likely to switch providers. Generally,
customer switching between competing suppliers is evidence that consumers have viable
competitive alternatives.45
46.
There is significant customer switching in the broadband industry, particularly in recent
years. This increase in consumer switching likely reflects the intensifying competition between
broadband providers, driven by the deployment of fiber networks, and the upgrading of
broadband networks by cable providers. For instance, a 2010 Commission study found that 36
43
Federal Communications Commission, National Broadband Plan, March 17, 2010 at 3.
44
The White House Office of Science and Technology Policy and the National Economic Council, “Four Years of
Broadband Growth,” June 2013 at 4-5.
45
However, the absence of switching does not necessarily imply an absence of competition. For instance, low
propensity to switch may be the result of customer loyalty or long-term contracts, which may be the result of
competition between firms rather than reflect the absence of competition.
20 | P a g e
percent of Internet users had switched broadband service providers in the prior three years.46 Of
those users that switched providers for reasons other than moving, 54 percent stated that getting a
better price was a “major reason” to switch, and 55 percent said a major reason for switching was
to get higher broadband speeds.47 A 2011 U.S. government study of Internet use by U.S.
consumers found that when consumers switched broadband providers, 38 percent did so because
of price, 30 percent to obtain faster broadband speeds, 10 percent because of reliability, and 7
percent because of customer service.48
47.
Switching between providers in response to competitive offerings is enhanced by the fact
that broadband consumers tend to be well informed. In the broadband industry, marketing and
advertising raise consumer awareness of broadband services, and allow consumers to compare
competitive offerings.49 Industry groups and publications (e.g., Consumer Reports, J.D. Power
and Associates) provide surveys and comparisons of broadband provider offerings, and report
user satisfaction with various providers, allowing consumers to make informed decisions when
choosing providers.50
48.
Information to consumers also is provided by content providers themselves. Some
content providers report the speed ratings of Internet service providers for delivery of the
provider’s content. For instance, Google launched a speed test tool for YouTube videos called
Google Video Quality Report, allowing users to compare YouTube video streaming speeds by
46
Federal Communications Commission, “Broadband Decisions: What Drives Consumers to Switch – or Stick With
– Their Broadband Internet Provider,” December 2010 at 2-3, available at
http://www.fcc.gov/encyclopedia/broadband-speed.
47
Federal Communications Commission, “Broadband Decisions: What Drives Consumers to Switch – or Stick With
– Their Broadband Internet Provider,” December 2010 at 9, available at
http://www.fcc.gov/encyclopedia/broadband-speed.
48
National Telecommunications and Information Administration and Economics and Statistics Administration,
“Exploring the Digital Nation: America’s Emerging Online Experience,” June 2013 at 23.
49
For example, Comcast maintains a website entitled “XFINITY vs. the Competition” that compares its broadband
and Wi-Fi services to those of other providers. (“XFINITY from Comcast vs. Verizon FiOS,” available at
http://www.comcast.com/compare/fios.html.) AT&T maintains a similar site. (“Compare AT&T U-verse to
Comcast,” available at http://compareuverse.att.com/LandingPage.html#!/zipcode/60515/provider/Comcast.)
50
See, e.g., “J.D. Power 2013 U.S. Residential Television Service Provider Satisfaction Study,” September 26,
2013, available at http://www.jdpower.com/press-releases/2013-us-residential-television-service-providersatisfaction-study; Consumer Reports, “Save a bundle - How to piece together a great deal for TV, phone, and
Internet service,” February 2010, available at http://www.consumerreports.org/cro/magazinearchive/2010/february/electronics-and-computers/bundling/overview/bundling-ov.htm.
21 | P a g e
their service provider versus the speeds of other service providers.51 A variety of web sites, as
well as many broadband providers themselves (including Verizon, Charter, and Comcast), offer
Internet speed test tools that allow users to test the speed of their broadband connection.52 This
information allows consumers to evaluate and choose the best supplier for their needs.
49.
To the extent that concerns exist regarding differential treatment of content provider
traffic by broadband providers, such behavior could be monitored by consumers. In fact, the
content providers themselves would have an incentive to inform consumers.53 And, the
Commission’s own transparency rules will further ensure that the information necessary to
evaluate the performance of broadband providers is available to consumers.54
D. Consumer benefits from broadband competition
50.
The competitive rivalry between wireline broadband access providers has led to falling
prices, higher broadband speeds, increased availability of high-speed broadband, greater
consumer choice of products and packages, and other improvements in quality.
51.
Lower quality-adjusted prices: U.S. consumer prices for wireline broadband have
decreased considerably over time. The decline in broadband prices is much more pronounced
51
Google Video Quality Report, available at http://www.google.com/get/videoqualityreport/; Angela Moscaritolo,
“Test Your ISP’s Video Quality With YouTube Tool,” PCMag, May 29, 2014, available at
http://www.pcmag.com/article2/0,2817,2458723,00.asp. The article states that “[p]roviders will receive a top-notch
‘HD Verified’ rating if they can ‘consistently deliver HD video, a resolution of at least 720p, without buffering or
interruptions,’” and notes that FiOS is an HD-verified provider.
52
See, e.g., CNET, Bandwidth Meter Speed Test, available at http://www.cnet.com/internet-speed-test/; Charter
Communications Speed Test, available at http://speedtest.charter.com/; Verizon Speed Test, available at
http://my.verizon.com/services/speedtest/?CMP=DMC-CVZ_ZZ_ZZ_Z_ZZ_N_Z00139.
53
This type of information dissemination has been observed in programming disputes between content providers
and MVPDs. For instance, in August 2013, Time Warner Cable and CBS entered a dispute over carriage fees for
CBS networks, as well as premium cable networks like Showtime, which resulted in a blackout of the channels for
Time Warner Cable subscribers in major metropolitan areas including New York, Los Angeles, Boston and
Chicago. Both Time Warner Cable and CBS released public statements and made posts via corporate sites and
social media during the month-long battle attacking the other. (Sarah Barry James, “Time Warner Cable-CBS
Blackout Begins,” SNL Kagan, August 2, 2013 available at
http://www.snl.com/InteractiveX/article.aspx?TabStates=0&id=18870079&KPLT=2.) CBS encouraged Time
Warner Cable subscribers to switch to other providers such as Verizon FiOS or DIRECTV in order to view CBS
programming. (David Lieberman, “No Deal! CBS and Showtime Go Dark On Time Warner Cable,” Deadline
Hollywood, August 2, 2013, available at http://www.deadline.com/2013/08/no-deal-cbs-goes-dark-on-time-warnercable/.)
54
Federal Communications Commission, Report and Order, Preserving the Open Internet Broadband Industry
Practices, GN Docket No. 09-191, December 23, 2010 (hereinafter, Open Internet Order), ¶¶ 53-61.
22 | P a g e
after accounting for increasing quality, such as higher speeds. In particular, consumer bandwidth
costs have declined an average of 27 percent annually over the period 1999 to 2013, falling from
$1,245 per Gbps in 1999 to $16 per Gbps in 2013.55
52.
Higher broadband speeds: U.S. broadband speeds have increased for virtually all
locations and users. From 2006 to 2013, the average speed (across all wireline providers)
offered on “standard packages” increased by at least 400 percent for every broadband technology
(FTTH, cable, DSL), as shown in Figure 6 below.56 As mentioned, some providers offer much
higher maximum speeds—Verizon’s FiOS network currently offers broadband download speeds
as high as 500 Mbps (up from 50 Mbps in 2007).
Figure 6: Average Wireline Broadband Speeds by Technology
160
Projection
Cable
140
DSL
FTTH
Average Download Speed (Mbps)
120
100
80
60
40
20
0
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
Notes: Based on Standard Package offered by providers. Cable and DSL speeds are the average of the Top and Bottom speeds listed in the data. DSL includes
AT&T U-verse.
Source: SNL Kagan 2013, "Media Trends: Actionable Metrics, Benchmarks & Projections For Major Media Sectors," July 2013 at 108, 110.
55
Mary Meeker, “Internet Trends 2014 – Code Conference,” KPCB, May 28, 2014 at 72.
56
SNL Kagan, “Media Trends: Actionable Metrics, Benchmarks & Projections for Major Media Sectors,”
December 2013 at 108, 110. The Commission recently reported that, for all broadband technologies, “average
subscribed speed is now 21.2 Mbps, representing an average annualized speed increase of about 36 percent.”
(Federal Communications Commission Office of Engineering and Technology and Consumer and Governmental
Affairs Bureau, “2014 Measuring Broadband America Fixed Broadband Report,” June 2014 at 13.)
23 | P a g e
53.
Looking ahead, SNL Kagan forecasts that maximum download wireline broadband
speeds in the U.S. will continue to increase over the next several years.57 Cisco recently
predicted that average wireline broadband speeds will more than double by 2018.58
54.
Broader availability of high-speed broadband: As a result of the historical and ongoing
investments by broadband providers, most consumers in the U.S. now have access to high speed
broadband services. Over 50 percent of consumers have access to broadband download speeds
of at least 100 Mbps, roughly 80 percent have access to speeds of over 50 Mbps, and over 90
percent have access to speeds over 10 Mbps (see Figure 7 below).
Figure 7: Share of U.S. Consumers by Available Wireline Broadband Download Speeds
50 Mbps to 100 Mbps
23.5%
25 Mbps to 50 Mbps
2.7%
Greater than 100 Mbps
56.7%
10 Mbps to 25 Mbps
8.5%
3 Mbps to 10 Mbps
2.6%
Up to 3 Mbps
6.1%
Note: Data as of June 30, 2013.
Source: FCC National Broadband Map, available at http://www.broadbandmap.gov/summarize/nationwide.
57
SNL Kagan, “Media Trends: Actionable Metrics, Benchmarks & Projections for Major Media Sectors,”
December 2013 at 108, 110.
58
“Cisco VNI Forecast Highlights,” Cisco, available at
http://www.cisco.com/web/solutions/sp/vni/vni_forecast_highlights/index.html.
24 | P a g e
55.
Greater consumer choice of packages and services: Broadband providers offer a wide
array of subscription plan choices to consumers, allowing users to pay for broadband access
plans with the particular performance characteristics they demand. For example, Verizon offers
FiOS plans with download speeds ranging from 15 to 500 Mbps. Comcast offers multiple highspeed data plans to its users, generally with speeds ranging from 3 Mbps to 105 Mbps.59
56.
Broadband providers are also offering subscribers enhanced services, such as control of
their home TV via smartphone and tablet applications. For instance, Verizon offers a My FiOS
application, which not only allows users to use their tablet or smartphone as a remote control, but
also allows for remote setting of their DVR and management of their Verizon account to pay
bills, listen to voicemail, review call logs and get technical support.60 Comcast’s Xfinity has a
mobile app that allows users to change channels on their TV, schedule DVR programming
remotely, search program listings, and watch on demand shows.61
E. The increasing role of interconnection arrangements and content delivery
networks (CDNs)
57.
Content providers are using various ways to bypass the Internet “backbone,” and
interconnection points between various networks and sources of Internet traffic, in order to
enable faster and more reliable content delivery. Networks and content providers are connecting
directly more often, and in more places.62 The exchange of traffic over these connections is
governed by an increasing variety of arrangements, such as “paid peering” (where one party
compensates another for connection and access to its network)63 and “quality of service”
59
SNL Kagan, “HSD Pricing by Provider,” 2013. Comcast also offers an Internet plan with a top download speed
of 505 Mbps in a handful of markets. (“Comcast Fights Fiber with Fiber,” Multichannel News, September 23,
2013.)
60
Verizon My FiOS, available at https://www.verizon.com/fiostv/myapps/unprotected/apps.aspx.
61
The Xfinity TV Remote App, available at http://xfinity.comcast.net/learn/internet/mobile-tv-app/.
62
See, e.g., Bret Swanson, “How the Net Works: A Brief History of Internet Interconnection,” Entropy Economics,
February 21, 2014 at 2.
63
Internet backbone providers always have “paid” each other for the exchange of traffic. While, historically,
“settlement free peering” existed, these arrangements occurred between networks with roughly equal traffic flows
across exchange points, such that traffic was the “currency” used by the parties. Like other bartering arrangements,
if the exchange became “asymmetric” (i.e., when one network began to deliver substantially more traffic than the
other), “paid peering” arrangements are used to compensate the party receiving more traffic. (Bret Swanson, “How
the Net Works: A Brief History of Internet Interconnection,” Entropy Economics, February 21, 2014 at 2.)
25 | P a g e
agreements (which guarantee a certain level of transit quality by the network provider, such as
low latency or packet loss).
58.
The exchange of traffic is increasingly facilitated by firms other than content providers or
broadband networks. CDNs, for example, sell services to content and application providers that
improve the speed and quality at which their content can be accessed by users. CDNs reduce the
number of potential congestion points for delivering content from providers’ data centers to end
users.64 For instance, content providers pay CDNs to cache frequently requested web content at
many locations throughout the Internet to enable faster content delivery and lower latency.
Many content providers use third party CDNs such as Akamai, Limelight, and Chinacache.
Moreover, very large content providers such as Netflix, Google, and Amazon have developed
their own proprietary CDNs.65
59.
These ways of increasing the efficiency of delivering products and services over the
Internet are becoming more and more common. For instance, CDNs are projected to carry over
half of all Internet traffic by 2018.66 With regard to online video, 67 percent of traffic is
projected to be delivered by CDNs by 2018, up from 53 percent in 2013.67 Video traffic is
forecast to be the primary driver of growth in CDN revenue in coming years, accounting for 81
percent of total CDN revenue by 2017.68
60.
CDNs, along with other interconnection arrangements, play an important role in
facilitating the efforts of content providers of all types and sizes to more effectively reach
64
See, e.g., Cisco, “Internet Video: New Revenue Opportunity for Telecommunications and Cable Providers,” July
2010. CDNs have been around since the mid-1990s, but their business models have evolved as the broadband
industry has changed. In addition to providing content caching, some CDNs may license their services to other
network operators. “Retail” CDNs specialize in “managed content” (IPTV and VOD), while other CDNs offer
support for over-the-top (OTT) content streaming and multi-device access (e.g., TV Everywhere). CDNs also
provide “value-added” services such as website and application acceleration, online security, and digital publishing.
(informa, “Content Delivery Networks: Market dynamics and growth perspectives,” 2012 at 4.)
65
See, e.g., Anna-Maria Kovacs, “Telecommunications competition: the infrastructure-investment race,” Internet
Innovation Alliance, October 8, 2013 at 7; Jon Brodkin, “Netflix’s many-pronged plan to eliminate video playback
problems,” Arstechnica, May 13, 2014, available at http://arstechnica.com/information-technology/2014/05/netflixsmany-pronged-plan-to-eliminate-video-playback-problems/.
66
Cisco, “Cisco Visual Networking Index: Forecast and Methodology, 2013–2018,” June 10, 2014 at 2.
67
Cisco, “Cisco Visual Networking Index: Forecast and Methodology, 2013–2018,” June 10, 2014 at 2.
68
informa, “Content Delivery Networks: Market dynamics and growth perspectives,” 2012 at 4.
26 | P a g e
users.69 Third-party CDNs allow content providers that do not have significant scale to improve
the speed and quality at which their content can be accessed by users, in competition with larger
providers that have the scale to invest in their own proprietary CDNs. Thus, third-party CDNs
“level the playing field” between large and small content providers in terms of ensuring the
efficient delivery of their content to users. CDNs also may provide a competitive alternative to
arrangements with broadband providers for both small and large content providers.
III.
Competition between Wireless Broadband Providers
A. Consumers have many options of mobile broadband providers
61.
There is vigorous competition in the wireless broadband industry. Consumers today can
obtain wireless broadband from many providers, with almost 90 percent of the U.S. population
having access to four or more wireless broadband carriers.70
62.
The majority of U.S. consumers also have access to high-speed 4G LTE services from
multiple providers. As one industry analyst noted:
Now that Verizon’s LTE deployment covers about 300 million Americans, that
number is closer to 95 percent. With AT&T’s LTE deployment reaching 270
million Americans by year-end-2013 and 300 million Americans by mid-2014,
that number will be even higher, assuming the two networks do not overlap
completely. And, of course, that means that at least 95 percent of Americans will
have access to at least two competing mobile wireless broadband networks within
a year. Once T-Mobile and Sprint complete their LTE buildout to the roughly 225
and 200 million Americans that they have, respectively, committed to cover,
about 60 percent-70 percent of Americans will have access to between three and
four LTE networks.71
63.
Many U.S. consumers also have access to and use broadband technology that combines
wireline and wireless elements. So-called “fixed wireless” broadband uses wireline to deliver
broadband service close to homes, and then uses small wireless antennas to provide the final
69
An analyst report about the CDN market noted that “CDNs could help smaller OTT players disrupt the content
hierarchy.” (informa, “Content Delivery Networks: Market dynamics and growth perspectives,” 2012 at 7.)
70
Federal Communications Commission National Broadband Map, available at
http://www.broadbandmap.gov/summarize/nationwide. Data as of June 30, 2013.
71
Anna-Maria Kovacs, “Telecommunications competition: the infrastructure-investment race,” Internet Innovation
Alliance, October 8, 2013 at 19.
27 | P a g e
connection to households. As the Commission has noted, “[f]ixed wireless technologies using
longer range directional equipment can provide broadband service in remote or sparsely
populated areas where other types of broadband would be too costly to provide.”72 The
Commission also has noted that fixed wireless broadband is used by over two million subscribers
in the U.S., mostly in rural areas.73
B. Firm behavior is consistent with significant competition
64.
Price competition: Wireless providers compete fiercely for customers on the basis of
price, and with respect to other important aspects of the wireless ecosystem, including the
provision of valuable services, handset devices, operating systems, applications, and content.
Competitive responses to rivals by wireless providers include reducing prices, dropping longterm contractual requirements, paying off early-termination fees, and offering monthly
installment plans for handset purchases.74
65.
Wireless broadband providers also have competed in recent years by providing diverse
pricing options. For instance, during 2011, providers were experimenting with various models of
usage-based pricing, with tiered pricing dependent on usage for each mobile device.75 In the
72
Federal Communications Commission, “Getting Broadband,” available at http://www.fcc.gov/guides/gettingbroadband.
73
Federal Communications Commission National Broadband Plan at 37; Robert. C. Atkinson, Ivy E. Schultz, Travis
Korte, and Timothy Krompinger, “Broadband in America – 2nd Edition, Where It Is and Where It Is Going, Update
of the 2009 Report Originally Prepared for the Staff of the FCC’s Omnibus Broadband Initiative,” May 2011 at 71.
74
Carriers’ actions in recent months exemplify the “tit-for-tat” competition that has characterized the wireless
broadband industry. In December 2013, AT&T introduced a new mobile sharing plan that offered discounts of $15
a month if subscribers were put on a Next plan (nonsubsidized device). Then in January 2014 Sprint announced
“Framily” (Friends and Family) plans that give subscribers discounts as they expand their “Framily.” The next
month, AT&T added an additional $10 per line discount to Next customers with at least a 10GB of data plan. At the
same time Verizon launched its “More Everything” plans that gave its Edge (nonsubsidized device) subscribers a
$10 per line discount on plans with less than 10GB of data and a $20 per line discount on Edge plans above 10GB
data. In March 2014, AT&T lowered the price of its 2GB plan for both its subsidized and Next subscribers, while
T-Mobile increased the allotted data for some of its plans by 500MB, and added an additional 5 Gb data tier. In
April 2014, Verizon implemented another $5 discount per line on its 10GB plus Edge plan. In the same month
Sprint launched an initiative where it would pay up to $350 in early termination fees and $300 for an old phone if a
subscriber switched to Sprint and enrolled in a Framily plan. (Bank of America Merrill Lynch, “1Q14: Preview &
Model Book – Everybody’s Doing It,” April 14, 2014 at 5.)
75
The Commission’s recent Mobile Wireless Competition Report describes that “three distinct models for
smartphone data pricing emerged in 2011: (1) tiered, usage-based data pricing with overage charges (Verizon and
AT&T); (2) tiered, usage-based data pricing with speed reductions instead of overage charges (T-Mobile); and (3)
unlimited data pricing (Sprint).” (Federal Communications Commission, 16th Annual Report and Analysis of
Competitive Market Conditions With Respect to Mobile Wireless, Including Commercial Mobile Services, March
28 | P a g e
latter part of 2012, providers offered consumers more freedom in regards to total wireless data
usage. Verizon and AT&T introduced “shared” data plans which allow consumers to “pool”
their data use across multiple devices on one plan, T-Mobile reintroduced an unlimited data
wireless plan option, and Sprint maintained its own unlimited data plan.76
66.
Many wireless service plans now allow subscribers the freedom to upgrade their devices
whenever they want instead of waiting for the end of a two-year contract.77 Recently, major
wireless providers have begun to offer non-subsidized device plans, where the customer pays full
price for the handset (either upfront or in interest free monthly installments) and then receives a
discount on their monthly service charge.78
67.
“Platform” competition: Competition among wireless broadband providers takes place
in an “ecosystem” that includes various complementary products and services, including wireless
devices, operating systems, applications, and other services. In this environment, wireless
broadband providers compete for customers by offering innovative broadband-connected mobile
devices (which provide access to specific mobile platforms, such as iOS, Android, or Windows
Mobile) and mobile services. Wireless carriers generally purchase handsets in bulk quantities
from smartphone manufacturers, and then sell the devices to consumers at a discount in
combination with a service plan.79
68.
The devices offered by wireless carriers, and the corresponding mobile platform accessed
by the devices, have become important dimensions of competition for wireless carriers. The
release of Apple’s iPhone (and the iOS platform) illustrates the importance of this competition.
21, 2013 (hereinafter “Federal Communications Commission, 16th Annual Mobile Wireless Competition Report”),
at 98.)
76
Federal Communications Commission, 16th Annual Mobile Wireless Competition Report at 98. Providers also
continued to offer low-priced, low-data usage plans for consumers with limited data needs, for instance. As the
Commission notes, “to continue attracting new customers with limited data needs, Verizon kept two entry-level
plans for new customers with basic phones.” (Federal Communications Commission, 16th Annual Mobile Wireless
Competition Report at 102-103.)
77
Brad Molen, “Your Carrier Wants You To Buy Into Early Upgrades, But Should You Take the Bait?” Engadget,
February 19, 2014, available at http://www.engadget.com/2014/02/19/early-upgrade-plans/.
78
See, e.g., Mark Rogowsky, “Verizon Tweaks Its Plans Again: Customers Win, Contracts Lose,” Forbes, April 15,
2014, available at http://www.forbes.com/sites/markrogowsky/2014/04/15/verizon-tweaks-its-plans-againcustomers-win-contracts-lose/.
79
Federal Communications Commission, 16th Annual Mobile Wireless Competition Report at 224.
29 | P a g e
When the iPhone was introduced in 2007 as an exclusive AT&T device, it was seen as the main
driver behind AT&T’s subscriber gains vis-à-vis other wireless carriers.80 Some analysts
estimated that AT&T paid more for the deal than it made back from its iPhone subscribers.81 To
compete with AT&T, Verizon Wireless partnered with Google and heavily promoted Androidbased phones made by Motorola and HTC. It was not until January 2011 that Verizon began
carrying the iPhone.82 Verizon’s share of the smartphone market increased 3.7 percentage points
after the introduction of the iPhone on its network.83 Sprint CEO Dan Hesse said Sprint’s lack of
iPhone carriage was the number one reason customers had been leaving Sprint.84 Sprint began
selling the iPhone in late 2011, and subsidized iPhone devices by as much as $450 per device. 85
In March 2013, T-Mobile launched a series of initiatives in order to better compete, headlined by
the mobile carrier finally offering the iPhone.86
80
Shayndi Raice and Yukari Iwantani Kane, “Verizon Finally Lands the iPhone,” The Wall Street Journal, January
8, 2011, available at http://online.wsj.com/news/articles/SB10001424052748704739504576068170230339348;
Peter Cohan, “Project Vogue: Inside Apple’s iPhone Deal with AT&T,” Forbes, September 10, 2013, available at
http://www.forbes.com/sites/petercohan/2013/09/10/project-vogue-inside-apples-iphone-deal-with-att/.
81
According to analyst Gene Munster of Piper Jaffray, AT&T paid Apple an estimated $18 per month per iPhone
subscriber, which exceeds the actual price of the iPhone over the two-year life of a user contract. (Tom Krazit,
“Piper Jaffray: AT&T paying Apple $18 per iPhone, per month,” CNET, October 24, 2007, available at
http://www.cnet.com/news/piper-jaffray-at-t-paying-apple-18-per-iphone-per-month/.)
82
Shayndi Raice and Yukari Iwantani Kane, “Verizon Finally Lands the iPhone,” The Wall Street Journal, January
8, 2011, available at http://online.wsj.com/news/articles/SB10001424052748704739504576068170230339348.
83
Verizon had a 26.7 percent share of U.S. smartphone subscribers before it began offering the iPhone (three month
average ending December 2010), and a 30.4 percent share a year later (three month average ending December
2011). (comScore MobiLens data; comScore, “U.S. Digital Year in Review,” February 2011 at 27, available at
http://www.cmpa.ca/sites/all/themes/cmpa/content/ind-publications/US-Digital-Year-in-Review-2010.pdf;
comScore, “2012 Mobile Future In Focus”, February 2012 at 15, available at
http://www.iab.net/media/file/comScore%2B2012%2BMobile%2BFuture%2Bin%2BFocus.pdf.)
84
David Goldman, “The iPhone Is A Nightmare For Carriers,” CNNMoney.com, February 8, 2012, available at
http://money.cnn.com/2012/02/08/technology/iphone_carrier_subsidy/.
85
Carriers such as Sprint pay Apple about $650 per iPhone. The cost to consumers is typically around $200, which
implies a “subsidy” of around $450 per device by the carrier. (Anton Troianovski and Don Clark, “Inside Nokia’s
Struggle Against Apple’s iPhone,” The Wall Street Journal, May 31, 2012, available at
http://online.wsj.com/news/articles/SB10001424052702303640104577436443119293310; based on cost for an
iPhone 4S 16-gigabyte device (see, Apple iPhone 4s (Sprint) Review
(http://www.pcmag.com/article2/0,2817,2394683,00.asp); JoAnn S. Lublin and Spencer E. Ante, “Inside Sprint’s
Bet on iPhone,” The Wall Street Journal, October 4, 2011, available at
http://online.wsj.com/news/articles/SB10001424052970203405504576603053795839250. See also, David
Goldman, “The iPhone Is A Nightmare For Carriers,” CNNMoney, February 8, 2012, available at
http://money.cnn.com/2012/02/08/technology/iphone_carrier_subsidy/.)
86
Marguerite Reardon, “T-Mobile Finally Gets the iPhone,” CNET, March 26, 2013, available at
http://www.cnet.com/news/t-mobile-finally-gets-the-iphone/.
30 | P a g e
69.
Investment competition: As with wireline networks, investments in upgrading wireless
networks in order to offer customers better-quality wireless broadband services is an important
competitive dimension. Wireless broadband providers have invested substantial amounts in their
wireless network infrastructure in recent years, including to deploy high-speed broadband
services to consumers and to improve the breadth of network coverage and capacity.87 In
particular, all major wireless carriers have rolled out so-called “fourth-generation” (“4G”)
technology, which allows the provision of much higher data speeds over wireless broadband.88
70.
Verizon was the first and leading wireless provider to roll out 4G in the U.S., and is
considered “the pioneer in LTE deployment.”89 Verizon began deploying its 4G LTE network in
late 2010, and aggressively expanded its 4G network coverage—by 2012 the majority of its data
traffic was traveling on 4G.90 AT&T, Sprint, and T-Mobile subsequently launched their 4G
networks (in 2011, 2012, and 2013, respectively).91 Verizon currently offers 4G LTE in over
500 cities, covering over 97 percent of the U.S. population. AT&T offers 4G coverage to 88
percent, Sprint to 71 percent, and T-Mobile to 63 percent.92 According to a recent Mobility
87
These investments include both cell sites and the backhaul connections between sites, which facilitate increased
speed and capacity. (Federal Communications Commission, 16th Annual Mobile Wireless Competition Report at
127.)
88
“4G” includes a variety of technological specifications. The terms “3G” and “4G” are used by industry for
marketing purposes, as well as by the International Telecommunications Union (ITU) for technical specifications.
For example, T-Mobile, AT&T, and Verizon Wireless refer to their WiMAX, HSPA+, and LTE networks as “4G.”
(Federal Communications Commission, 16th Annual Mobile Wireless Competition Report at 123.)
89
Marguerite Reardon, “T-Mobile launches 4G LTE network,” CNET, March 26, 2013, available at
http://www.cnet.com/news/t-mobile-launches-4g-lte-network/.
90
Marguerite Reardon, “Verizon: Our 4G LTE network will soon carry most of our data,” CNET, October 9, 2012,
available at http://www.cnet.com/news/verizon-our-4g-lte-network-will-soon-carry-most-of-our-data/.
91
Phil Goldstein, “AT&T to launch LTE Sunday, September 18,” FierceWireless, September 15, 2011, available at
http://www.fiercewireless.com/story/att-launch-lte-sunday/2011-09-15. Marguerite Reardon, “Sprint officially
launches 4G LTE in 15 cities,” CNET, July 16, 2012, available at http://www.cnet.com/news/sprint-officiallylaunches-4g-lte-in-15-cities/; Marguerite Reardon, “T-Mobile launches 4G LTE network,” CNET, March 26, 2013,
available at http://www.cnet.com/news/t-mobile-launches-4g-lte-network/.
92
Verizon Communications 2013 Annual Report at 10, available at
http://www.verizon.com/investor/annualreports.htm; AT&T Inc. 2013 Annual Report at 5, available at
http://www.att.com/gen/investor-relations?pid=9186; The Sprint Quarterly Investor Update, April 29, 2014 at 3,
available at
http://investors.sprint.com/Cache/1001186417.PDF?Y=&O=PDF&D=&fid=1001186417&T=&iid=4057219; TMobile 2013 Annual Report at 4-5, available at http://investor.tmobile.com/GenPage.aspx?IID=4091145&GKP=1073748984; US Census Bureau Population Clock, December 31,
2013, available at http://www.census.gov/popclock/.
31 | P a g e
Report by Ericsson, by 2019, 85 percent of mobile subscriptions in North America will be LTE
(a particular type of 4G technology).93
71.
Verizon recently began further upgrades of its 4G network. In particular, Verizon’s 4G
“XLTE” delivers faster peak data speeds, and double the bandwidth compared to “regular” 4G.94
The technology is expected to improve performance on Verizon’s wireless broadband network,
especially in densely populated areas.95 As of June 2014, the XLTE 4G network had been
launched in over 300 of Verizon’s 500 4G LTE-ready cities.96
72.
Investments by wireless providers in 4G and other wireless technology and infrastructure
continue to be sizeable, and are only expected to increase. Since 2002, U.S. wireless providers
have invested on average a combined $22.6 billion per year to upgrade their networks.97 More
recently, annual investment in U.S. wireless networks grew from $21 billion in 2009 to $30
billion in 2012.98 In 2011, wireless broadband providers Verizon and AT&T were the top two
American firms in terms of capital expenditures.99
C. Consumer behavior is consistent with significant competition
73.
As with wireline broadband, consumers evaluate competing offers from multiple wireless
broadband providers, and choose the optimal device and plan for their location, data needs, price
93
Ericsson, “Ericsson Mobility Report On the Pulse of the Networked Society,” June 2014 at 9, available at
http://www.ericsson.com/news/1790097?categoryFilter=press-releases_1270673222_c.
94
Debi Lewis, “XLTE: America’s Best Network Gets Even Better,” VerizonWireless, June 26, 2014 available at
http://www.verizonwireless.com/news/article/2014/05/verizon-wireless-xlte.html. As of June 2014, Verizon was
continuing its rollout of its XLTE network improvements. XLTE “delivers faster peak data speeds and a minimum
of double the bandwidth to 4G LTE customers in high traffic areas in markets nationwide where Advanced Wireless
Services (AWS) spectrum has been activated.” The technology allows capable devices to access both 700 MHZ and
AWS spectrum, and thus also frees up bandwidth for non-XLTE ready devices on the 700 MHz spectrum.
95
Angela Moscaritolo, “Verizon Brings Super-Charged XLTE to 300 Markets,” PCMag, June 27, 2014, available at
http://www.pcmag.com/article2/0,2817,2460175,00.asp.
96
Angela Moscaritolo, “Verizon Brings Super-Charged XLTE to 300 Markets,” PCMag, June 27, 2014, available at
http://www.pcmag.com/article2/0,2817,2460175,00.asp;
http://s7.vzw.com/is/content/VerizonWireless/eCatalogs/Verizon-XLTE-markets.pdf.
97
See Figure 13 infra.
98
The White House Office of Science and Technology Policy and the National Economic Council, “Four Years of
Broadband Growth,” June 2013 at 2.
99
Excludes R&D. Diana Carew and Michael Mandel, “Investment Heroes: Who’s Betting on America’s Future?”
Progressive Policy Institute, July 2012 at 3.
32 | P a g e
range, and other factors. The significant switching by wireless customers indicates the
competitive rivalry in the wireless industry. For instance, average wireless subscriber monthly
churn rates have been between 2.0 percent and 2.5 percent since at least 2005.100 This means
that about 25 percent of customers churn in a year.101
74.
Mobile broadband price, data coverage, and download speed are the top reasons why
consumers switch providers. An industry study found that “after price, the biggest reason
smartphone owners switched operators was to get better data coverage and download speeds. In
fact, 40 percent said they switched operators in the past year to get better data speed and
coverage compared with 26 percent who said they switched to get better voice coverage.”102
Another recent survey found that 45 percent of mobile users would be likely to switch wireless
carriers to obtain better coverage.103
75.
Although wireless subscribers sometimes enter into multi-year contracts, several
countervailing factors work to reduce switching costs. For example, major wireless providers
pro-rate their early termination fees (“ETFs”), making switching less costly for consumers under
long-term contracts.104 In fact, some competitors, like T-Mobile and Sprint, pay off consumers’
ETFs if users switch to their service, which means that consumers can switch at no cost.105
Moreover, “secondary markets” for mobile contracts and devices facilitate switching/cancelling
of contractual obligations undertaken by users, as the Commission has noted.106
100
Federal Communications Commission, 16th Annual Mobile Wireless Competition Report at 165.
101
Although some wireline churn results from people moving from an area served by their current provider to an
area where that provider does not operate, most wireless broadband suppliers operate nationwide.
102
Sue Marek, “Study: Data speed is more critical than voice coverage for smartphone users,” FierceWireless, July
13, 2012, available at http://www.fiercewireless.com/story/study-data-speed-more-critical-voice-coveragesmartphone-users/2012-07-13?utm_medium=nl&utm_source=internal.
103
Stuart Taylor, “Service Provider Wi-Fi and Small Cells: Discover What Consumers Want from Wi-Fi and
Mobile,” Cisco, 2013 at 8.
104
Federal Communications Commission, 16th Annual Mobile Wireless Competition Report at 120.
105
“T-Mobile Delivers Contract Freedom for Families By Paying Off Early Termination Fees,” January 8, 2014,
available at http://newsroom.t-mobile.com/news/t-mobile-delivers-contract-freedom-for-families-by-paying-offearly-termination-fees.htm; “For a Limited Time, Customers Who Switch a Number to Sprint on a Family Plan Can
Save up to $650,” April 4, 2014, available at http://newsroom.sprint.com/news-releases/for-a-limited-timecustomers-who-switch-a-number-to-sprint-on-a-framilysm-plan-can-save-up-to-650.htm.
106
Federal Communications Commission, 16th Mobile Competition Report at 121: “The emergence of a secondary
market segment for mobile wireless service contracts may facilitate consumers’ ability to switch service providers.
33 | P a g e
76.
Wireless broadband consumers have access to information about rival offerings and make
informed decisions between competing wireless broadband providers. This information is
available from marketing and advertising by wireless broadband providers, consumer groups,
trade associations, other third parties, and government agencies. As the Commission has noted:
In order to make informed decisions, consumers need detailed information about
the price, availability, quality, and features of mobile wireless services. All
mobile wireless service providers offer resources on their websites that advertise
their products, services, and prices and that give potential customers information
on their networks, service plans, and terms of service. A number of third parties –
such as Consumer Reports, trade associations, marketing and consulting firms,
and several websites – also provide consumers with an overview and comparison
of the mobile wireless services available in their local areas. In addition,
organizations such as Consumer Reports and J.D. Power publish the results of
their wireless user surveys, which rate wireless service providers based on
customer satisfaction. … Information on mobile broadband availability can also
be found in the National Broadband Map.107
D. Consumer benefits from wireless broadband competition
77.
Competition between wireless providers has generated important consumer benefits.
Mobile wireless speeds continue to rise, prices per megabyte of data continue to fall, and
consumers have access to a wide and increasing variety of devices and plans. There is no
indication that this competitive rivalry will diminish, or that the expanding consumer benefits
generated by this competition will subside.
78.
Lower quality-adjusted prices: Quality-adjusted prices for wireless broadband (such as
prices per megabyte of data) have declined considerably in recent years. In the past few years,
average prices for wireless data services have declined from $0.46 per megabyte in 2008 to
$0.03 per megabyte in 2012, as shown in Figure 8 below.
… In addition to the secondary market for cellphone service contracts, there is a secondary market for iPhones and
other high-end smartphones and devices.”
107
Federal Communications Commission, 16th Annual Mobile Wireless Competition Report at 203-204.
34 | P a g e
Figure 8: Price of Wireless Broadband
$0.50
$0.46
$0.45
$0.40
$0.35
Price per MB
$0.30
$0.25
$0.20
$0.19
$0.15
$0.10
$0.10
$0.06
$0.05
$0.03
$0.00
2008
2009
2010
2011
2012
Source: Visage, "Infographic: The Staggeringly Hugh Future of Mobility," September 6, 2012 available at
http://visagemobile.com/mobilityblog/2012/09/06/infographic-the-staggeringly-huge-future-of-mobility/.
79.
Greater choice of wireless devices, applications, and services: Wireless competition has
facilitated the availability of a wide variety of devices (and associated operating systems),
applications, and services that are complements to a robust wireless broadband ecosystem.108
Each segment of the mobile wireless “ecosystem” is highly competitive, with significant
investment and innovation by existing firms and new entrants. Competition and innovation in
each of these segments has provided consumers with a wide array of innovative products and
services. For instance, competition has led to the availability of a wide variety of sophisticated
108
Federal Communications Commission, 15th Annual Report and Analysis of Competitive Market Conditions With
Respect to Mobile Wireless, Including Commercial Mobile Services, June 27, 2011 at 88: “In addition to network
quality and advertising, a third component of non-price rivalry among mobile wireless service providers is the
differentiation of the downstream products that they offer or that rely on their networks, including handsets/devices,
operating systems, and mobile applications. . . . As mobile operating systems, and the functionalities and application
stores they enable, play a more prominent role in a consumer’s mobile wireless experience, consumers are showing
an increasing loyalty to particular operating systems or device platforms.”
35 | P a g e
mobile devices.109 As the Commission noted, “[f]rom 2006 to 2012, the number of mobile
wireless handset manufacturers that distribute in the U.S. market increased from eight to 23.
During June 2012, these 23 handset manufacturers offered a total of 266 handset models to
mobile wireless service providers in the United States.”110
80.
The availability of content, applications, and services customized for a mobile
environment has grown exponentially. Mobile applications (“apps”), in particular, have
proliferated.111 Apple’s App Store and Google Play currently both offer over one million apps
for consumers to download for iOS and Android mobile devices respectively.112 Increased
wireless bandwidth offered by wireless carriers has facilitated the delivery of this mobile content.
81.
Higher broadband speeds and availability of high-speed broadband: Average wireless
broadband speeds have increased considerably in the past few years. 4G LTE network providers
typically offer average download speeds of 5 to 12 Mbps,113 which makes them comparable to or
faster than most wireline DSL services, although generally slower than wireline services using
fiber or DOCSIS 3.0 technologies. North America’s average wireless data connection speed is
the fastest in the world, nearly twice that available in Western Europe, and over five times the
109
The Commission has noted that “mobile wireless providers continue to compete by offering consumers a variety
of different mobile wireless devices with innovative features,” and that “price rivalry” between competing wireless
broadband providers takes the form “developments in mobile service pricing plans.” The Commission “report[ed]
evidence of significant actions, changes, and expenditures” in regards to providers’ competitive rivalry. (Federal
Communications Commission, 16th Annual Mobile Wireless Competition Report at 97-98, 145.)
110
Federal Communications Commission, 16th Annual Mobile Wireless Competition Report at 216. The majority of
these devices are smartphones with broadband capability. In 2013, smartphone unit sales accounted for 73 percent
of total wireless handset unit sales. (Telecommunications Industry Association, “TIA’s 2014-2017 ICT Market
Review and Forecast,” 2014 at 5-16.)
111
Federal Communications Commission, 16th Annual Mobile Wireless Competition Report at 24.
112
Nathan Ingraham, “Apple Announces 1 million apps in the App store, more than 1 billion songs played on iTunes
radio,” The Verge, October 22, 2013, available at http://www.theverge.com/2013/10/22/4866302/apple-announces1-million-apps-in-the-app-store; Josh Ownby, “Google Play Store Hits Million – Apps Milestone,” Gazelle, August
14, 2013, available at http://www.gazelle.com/thehorn/2013/08/14/google-play-store-hits-million-apps-milestone.
113
For instance, Verizon’s 4G LTE network offers download speeds of 5 to 12 Mbps.
(http://www.verizonwireless.com/insiders-guide/network-and-plans/4g-lte-speeds-compared-to-home-network/.)
36 | P a g e
global average.114 Cisco has recently projected that average North American wireless broadband
speeds will continue to increase rapidly, and will more than double by 2018.115
82.
Higher broadband speeds are available to the vast majority of American consumers. In
mid-2012, as 4G services were being rolled out, 79 percent of Americans had access to wireless
broadband speeds of 10 Mbps or higher.116 By mid-2013, the share of Americans with access to
wireless broadband download speeds of greater than 10 Mbps had increased to over 95 percent,
as shown in Figure 9 below. Only about 2 percent of consumers did not have access to
download speeds of at least 3 Mbps.
Figure 9: Share of U.S. Consumers by Available Wireless Broadband Download Speeds
3 Mbps to 10 Mbps
2.8%
Up to 3 Mbps
2.1%
10 Mbps to 25 Mbps
88.7%
Greater than 100 Mbps
3.0%
50 Mbps to 100 Mbps
1.4%
25 Mbps to 50 Mbps
2.0%
Note: Data as of June 30, 2013.
Source: FCC National Broadband Map, available at http://www.broadbandmap.gov/summarize/nationwide.
114
The White House Office of Science and Technology Policy and the National Economic Council, “Four Years of
Broadband Growth,” June 2013 at 2. Data for 2012.
115
Cisco, “The Zettabyte Era: Trends and Analysis,” June 10, 2014, Table 5.
116
Anna-Maria Kovacs, “Telecommunications competition: the infrastructure-investment race,” Internet Innovation
Alliance, October 8, 2013 at 19.
37 | P a g e
E. Increasing substitution between wireline and wireless broadband
83.
While the previous sections discussed competition among wireline providers separately
from competition among wireless providers, the lines between these traditional broadband
network distinctions have been blurring, and are expected to continue to blur. Consumers
increasingly connect to wireline and wireless networks for many of the same uses, and some
consumers rely exclusively on wireless for broadband. These changes in broadband usage have
been spurred by critical industry developments.
84.
Innovation and investments in both wireless broadband networks and wireless devices
increasingly have made wireless networks more competitive to wired networks, particularly to
lower-speed wireline options. First, significant improvements in speed and capacity of wireless
broadband networks have allowed consumers to perform many of the same tasks on wireless
devices as they perform on computers connected via wireline networks. Although generally
slower than wireline broadband options such as fiber and DOCSIS 3.0, 4G LTE network
providers typically offer average download speeds of 5 to 12 Mbps that are sufficient for many
uses and are comparable to or faster than traditional copper-based DSL services.117 As wireless
broadband providers continue to improve their networks to increase speeds,118 and as more
wireless spectrum becomes available,119 wireless networks are likely to become an alternative to
wireline providers for a larger range of broadband users and uses. Second, the advent of
advanced wireless devices, particularly tablets, has blurred the distinction between wireless and
wireline devices, and how content providers access consumers using those devices.
85.
As a result, although wireless networks do not offer the same speeds as next-generation
wireline networks like cable and fiber, wireless broadband services are to some extent a
competitive alternative for wireline networks for some consumers. Mobile-only broadband
117
For instance, Verizon’s 4G LTE network offers download speeds of 5 to 12 Mbps.
(http://www.verizonwireless.com/insiders-guide/network-and-plans/4g-lte-speeds-compared-to-home-network/.)
118
For example, recent tests show that Verizon’s 4G LTE service can provide between 12.7 Mbps and 53.7 Mbps.
(Daniel P. Howley, “Verizon Wins NYC 4G Showdown, Sprint Dead Last,” Laptop Magazine, March 14, 2014,
available at http://blog.laptopmag.com/fastest-4g-nyc.)
119
The Commission’s upcoming spectrum auctions will allocate new spectrum bands to wireless use. See, e.g.,
“FCC Adopts Rules For First Ever Incentive Auction; Will Make Available Additional Airwaves, Increase
Competition for Mobile Broadband,” Federal Communications Commission News Release, May 15, 2014, available
at http://transition.fcc.gov/Daily_Releases/Daily_Business/2014/db0515/DOC-327100A1.pdf.
38 | P a g e
households now account for an increasing share of total broadband connections. In particular,
there were 8.5 million mobile-only broadband households in the U.S. in 2013, accounting for 8.9
percent of total broadband connections.120 And, even consumers that maintain a wireline
connection are increasingly performing some tasks over wireless broadband connections that
previously had been done exclusively via a wireline connection. These tasks include web
browsing, shopping,121 streaming video,122 and various other activities that previously were
conducted primarily on PCs over wireline networks.123
86.
These facts indicate that consumers are growing increasingly used to performing tasks on
multiple broadband platforms, and wireless networks are becoming a competitive alternative to
wireline broadband for some consumers. Although the wireless broadband industry is still
nascent, and evolving rapidly, these industry developments suggest that wireless networks exert
some degree of competitive pressure on wireline broadband providers. With continued
innovation and investment, there is the opportunity for wireless broadband to be a more
attractive competitive alternative over time.
87.
The proliferation of Wi-Fi networks further demonstrates this demand-side
substitutability—in this case, substitution of wireline networks for wireless broadband access.
120
Strategy Analytics, “Service Provider Strategies (SPS), North America Broadband Forecast,” April 2014 at 3.
121
Nielsen reports that 87 percent of consumers with smartphones or tablets use those devices for shopping or
related activities, including researching their purchase, locating a store, comparing prices, and purchasing products
or services. (Bill Siwicki, “Consumer Internet use shifts from PCs to smartphones,” Internet Retailer, February 20,
2014, available at http://www.internetretailer.com/2014/02/20/consumer-internet-use-shifts-pcs-smartphones.)
122
Nielsen reported that in 2013, over 50 million people were watching video on their mobile devices, up from 25
million in 2010. (“Q&A with Nielsen on the State of the Mobile Industry,” Heartland Mobile Council, September 4,
2013 available at http://heartlandmobilecouncil.org/summit/qa-with-nielsen-on-the-state-of-the-mobile-industry/;
“Number of Americans Watching Mobile Video Grows More Than 40% in Last Year,” Nielsen, March 30, 2011,
available at http://www.nielsen.com/us/en/newswire/2011/number-of-americans-watching-mobile-video-growsmore-than-40-in-last-year.html.) Nielsen reported that the share of Netflix and Hulu users that streamed video on
their mobile phones and tablets increased eight-fold from 2011 to 2013. (“Binging Is the New Viewing for Overthe-top Streamers,” Nielsen, September 18, 2013; “What Netflix and Hulu Users Are Watching… And How,”
Nielsen, July 27, 2011.)
123
For instance, U.S. consumers now spend more time on the Facebook mobile app than spent accessing the
Facebook website via a computer. (“The Digital Consumer,” Nielsen, February 2014 at 19.) Consumers appear to
be substituting between wireless and wireline networks even for tasks conducted in the same place and/or time of
day. A recent McKinsey survey found that “[e]ven when a PC is nearby, almost a quarter of smart phone owners
say they prefer using their phone for email, and 23% for Internet browsing.” (Bertil Chappius, Ewan Duncan,
Brendan Gaffey and Kevin Roche, “The next stage: Six ways the digital consumer is changing,” McKinsey
iConsumer research, April 2012 at 3.)
39 | P a g e
To put it another way, wireline services combined with Wi-Fi is a competitive alternative to the
use of wireless networks for some customers and uses. Wi-Fi networks are ubiquitous not just at
home but also at the workplace, coffee shops, airports, businesses, parks, and entire cities,124 and
broadband providers are rapidly deploying “homespot” Wi-Fi networks.125
Wireless carriers
are also currently testing HotSpot 2.0, a technology that allows for seamless transition from WiFi to wireless broadband signal for users with automated authentication. 126 A significant share
of broadband use from mobile devices now takes place through Wi-Fi connected to a wireline
network rather than wireless networks.127
IV.
The Future Outlook for Broadband
88.
As described above, there has been staggering investments by broadband providers, and
the competitive successes of the broadband industry to date have benefited consumers greatly.
Looking forward, there are significant opportunities and challenges, as broadband providers
compete to bring innovative products and services to users. The rapid growth in demand for
broadband of the past few years, both on wireline and wireless networks, is expected to continue,
if not accelerate. This growth in demand for broadband to a large extent will be driven by
increasing demand for video content, both over wireline and wireless networks. In order to meet
124
See, e.g., Stuart Taylor, “Service Provider Wi-Fi and Small Cells: Discover What Consumers Want from Wi-Fi
and Mobile,” Cisco, 2013 at 2-3.
125
Homespots are residential Internet gateways (such as a cable box or router) that are enabled as semi-public Wi-Fi
access points for other subscribers of a broadband provider. Comcast alone plans to have 8 million Wi-Fi hotspots
across the United States by the end of 2014. (Bob Fernandez, “Comcast Aims for 8 Million WiFi Hotspots,”
Pilly.com, May 2, 2014, available at http://articles.philly.com/2014-05-02/business/49555320_1_comcast-corpnational-wireless-network-neil-smit.) Providers also have created Wi-Fi roaming agreements linking their hotspots.
For example, the U.S. Cable Wi-Fi Consortium joins providers Comcast, Time Warner Cable, Cox, Brighthouse and
Optimum, which have linked more than 200,000 hotspots. (Bank of America Merrill Lynch, “Global Wireless
Matrix 1Q14, The Quad Play: How Disruptive?” April 21, 2014 at 7.)
126
Bank of America Merrill Lynch, “Global Wireless Matrix 1Q14, The Quad Play: How Disruptive?” April 21,
2014 at 7. Apple’s latest iOS for iPhone uses HotSpot 2.0 technology.
127
About two-thirds of tablet users in the U.S. connect only via Wi-Fi. (Federal Communications Commission, 16th
Annual Mobile Wireless Competition Report at 238.) Another survey, conducted in 2013, found that about 80
percent of respondents connected their tablets, laptops, and e-Readers to the Internet exclusively through Wi-Fi.
(Stuart Taylor, “Service Provider Wi-Fi and Small Cells: Discover What Consumers Want from Wi-Fi and Mobile,”
Cisco, 2013 at 3.) Overall, more than half of data traffic from wireless devices is carried over Wi-Fi networks, and
this share is expected to increase to 64 percent by 2018. (“57 Percent of Wireless Data Traffic Carried Over Wi-Fi,”
CTIA, May 15, 2014, available at http://www.ctia.org/resource-library/facts-and-infographics/archive/data-over-wifi.)
40 | P a g e
this expected growth in consumer demand, and the changing broadband traffic patterns, wireless
and wireline broadband providers are expected to need to continue making significant
investments in network infrastructure.
A. Continued growth of consumer demand for broadband
89.
Demand for wireline broadband networks: There has been rapid growth in consumer
demand for high-speed broadband access in the past few years. Demand for wireline broadband
has risen steadily, as shown in Figure 10 below. In particular, between 2008 and 2014, wireline
broadband traffic in North America increased from about 2,600 petabytes per month to almost
20,000 petabytes per month.128
Figure 10: North America Wireline Broadband Traffic
25,000
IP Traffic (PB per month)
20,000
15,000
10,000
5,000
0
2008
2009
2010
2011
2012
2013
2014
Note: 2014 data forecasted as of June 10, 2014.
Sources: Cisco, “Cisco Visual Networking Index: Forecast and Methodology” Reports: 2007–2012 (June 16, 2008), Table 1; 2008–2013 (June 9, 2009), Table 1;
2009–2014 (June 2, 2010), Table 3; 2010–2015 (June 1, 2011), Table 2; 2011–2016 (May 30, 2012), Table 1; 2012–2017 (May 29, 2013), Table 1; 2013–2018
(June 10, 2014), Table 1.
128
The growth in demand for broadband is both due to an increase in the number of broadband subscribers and
increase in the use per subscriber. For instance, average monthly use per subscriber of wireline broadband has more
than doubled from the first half of 2011 to the second half of 2013. (Parks Associates, “New Trends in Digital
Delivery and CDNs,” January 2014 at iv.)
41 | P a g e
90.
Consumer demand for wireline broadband is expected to continue to grow rapidly, as is
the consumer demand for higher-speed wireline broadband services. For example, U.S. wireline
broadband traffic is expected to double between 2013 and 2018.129
91.
Demand for wireless broadband networks: The growth in consumer demand for
wireless broadband has been even more staggering in the past few years, as shown in Figure 11
below. The launch of the iPhone in June 2007 and the subsequent growth of smartphones have
led to an explosion in mobile broadband usage, as mobile Internet-connected devices
proliferated.130
The deployment to 4G technology also has contributed to the growth in demand
for wireless broadband, as 4G LTE devices sold in 2013 used 40 percent more bandwidth than
the 3G devices sold a year prior.131
129
“Cisco VNI Forecast Highlights,” Cisco available at
http://www.cisco.com/web/solutions/sp/vni/vni_forecast_highlights/index.html.
130
The Commission has noted: “Disruptive technology transformations happen once every 10 to 15 years. Mobile
broadband represents the convergence of the last two great disruptive technologies – Internet computing and mobile
communications – and may be more transformative than either of these previous breakthroughs.” (Federal
Communications Commission, “Connecting America: The National Broadband Plan,” March 16, 2010 at 75.)
131
Telecommunications Industry Association, “TIA’s 2014-2017 ICT Market Review and Forecast,” 2014 at 5-2.
42 | P a g e
Figure 11: North America Wireless Broadband Traffic
700
600
IP Traffic (PB per month)
500
400
300
200
100
0
2008
2009
2010
2011
2012
2013
2014
Note: 2014 data forecasted as of February 5, 2014.
Sources: Cisco, “Cisco Visual Networking Index: Global Mobile Data Traffic Forecast Update” Reports: 2008-2013 (January 29, 2009), Table 1; 2009-2014
(February 9, 2010), Table 7; 2010–2015 (February 1, 2011), Table 9; 2011–2016 (February 14, 2012), Table 5; 2012–2017 (February 6, 2013), Table 6; 2013–
2018 (February 5, 2014), Table 6.
92.
The rapid growth in consumer demand for wireless broadband is expected to continue for
the foreseeable future, which will require significant investments and innovation by wireless
broadband providers and various other industry participants. For example, wireless broadband
traffic is expected to increase by a factor of eight between 2013 and 2018 in the United States,
from 360 petabytes per month to 2,700 petabytes per month.132 This growth will create
significant challenges for wireless providers due to the significant capacity constraints that they
face.133
132
“Cisco VNI Forecast Highlights,” Cisco available at
http://www.cisco.com/web/solutions/sp/vni/vni_forecast_highlights/index.html.
133
See, e.g., Telecommunications Industry Association, “TIA’s 2014-2017 ICT Market Review and Forecast,” 2014
at 5-14, 5-15: “While LTE rollouts have helped accommodate the current demand for data, mobile network
operators are already facing capacity issues with their current spectrum holdings. Meanwhile, data traffic continues
to surge. The volume of data traffic is expected to be 10 times greater in 2018 than in 2013. Those traffic volumes
will require a next-generation or 5G network. Near instantaneous video downloading, sub-millisecond latency and
virtual overlay are among the features a 5G network hopes to deliver. … major efforts are taking place to begin the
process of identifying the technologies for 5G.”
43 | P a g e
93.
The growth in demand for broadband services, both on wireline and wireless networks,
will be driven largely by bandwidth-intensive applications and services, such as video. The
bandwidth used by different applications varies tremendously. Some consumer uses, such as email, impose very little bandwidth cost, while other uses, such as video delivered by over-the-top
(“OTT”) services, are much more bandwidth-intensive. Table 1 below shows the share of
wireline broadband downstream traffic by content provider. The top two providers in terms of
bandwidth use, Netflix and YouTube, account for over 47 percent of traffic. The top eight
providers account for almost two-thirds of the total wireline broadband Internet traffic, with
seven of these eight providers offering video services.134
Table 1: Top Wireline Bandwidth Using Applications (2014)
Rank
Application
Type of Application
1
2
3
4
5
6
7
8
Netflix
YouTube
iTunes
BitTorrent
MPEG
Facebook
Amazon Video
Hulu
Total
Video
Video
Video/Music
Video/Music
Video
Social Media
Video
Video
Share of
Downstream Traffic
34.2%
13.2%
3.6%
3.4%
2.9%
2.0%
1.9%
1.7%
62.9%
Notes: Data for North America, 2014. Based on peak bandwidth use period.
Source: Sandvine, "Global Internet Phenomena Report 1H 2014," May 15, 2014, Table 2.
94.
Video services will continue to drive demand for bandwidth.135 Cisco predicts that the
share of Internet traffic from video providers (including wireless and wireline in aggregate) will
grow from 66 percent in 2013 to 79 percent in 2018.136 In fact, more and more consumers are
134
Moreover, these specific uses are increasingly occurring at particular times of day. The top two content providers
in terms of bandwidth usage, Netflix and YouTube, accounted for 51 percent of broadband traffic at peak times on
wireline broadband networks. (Parks Associates, “New Trends in Digital Delivery and CDNs,” January 2014,
Figure 4.)
135
SNL Kagan, “Media Trends Actionable Metrics, Benchmarks & Projections For Major Media Sectors,”
December 2013 at 54.
136
Cisco, “Cisco VNI: Forecast and Methodology, 2013-2018,” June 10, 2014 at 2.
44 | P a g e
exclusively viewing video via broadband networks. SNL Kagan estimates such “cord-cutting”
could grow from 4.7 million households at the end of 2012 to nearly 13 million households by
2017.137 Moreover, ultra-high definition (UHD) video streaming, which has a 18 Mbps bit rate
(more than 9 times that of standard definition video streaming), is expected to further increase
demand for broadband bandwidth.138
95.
Video usage also will account for most broadband traffic on wireless networks. In the
next few years, the volume of mobile video traffic is expected to increase tenfold, and more than
half of all mobile data traffic is projected to come from video.139 These trends will create even
greater challenges for wireless providers due to the significant capacity constraints that they face.
96.
The growth in demand for broadband in the next few years also will be driven by the
“Internet of Things,” which includes broadband-connected appliances, cars, wearable devices,
etc., essentially involving machine-to-machine broadband use.140 Gartner estimates that by 2020
there will be 26 billion broadband-connected devices globally, and this will have a considerable
impact on data centers and network infrastructure.141 Cisco expects the number of broadbandconnected wearable devices (e.g., Google Glass, smart watches, fitness monitors) to grow
globally from 22 to 177 million during the same period.142 The growth of these connected
devices will require investments in infrastructure and innovative solutions to handle them.143
137
SNL Kagan, “Media Trends Actionable Metrics, Benchmarks & Projections For Major Media Sectors,”
December 2013 at 10.
138
By 2018, Cisco predicts that UHD will account for 22 percent of VOD broadband traffic worldwide. (Cisco,
“The Zettabyte Era: Trends and Analysis,” June 10, 2014, Figure 6.)
139
It is projected that the volume of mobile video traffic in 2017 will be nine times the volume in 2012. (Federal
Communications Commission, 16th Annual Mobile Wireless Competition Report at 173.) By 2019, more than 50
percent of global mobile data traffic is expected to come from video, up from about 40 percent in 2013. (Ericsson,
“Ericsson Mobility Report On the Pulse of the Networked Society,” June 2014 at 13, available at
http://www.ericsson.com/news/1790097?categoryFilter=press-releases_1270673222_c.)
140
See, e.g., Telecommunications Industry Association, “TIA’s 2014-2017 ICT Market Review and Forecast,” 2014
at 2-8.
141
“Gartner Says the Internet of Things Will Transform the Data Center,” Gartner Press Release, March 19, 2014,
available at http://www.gartner.com/newsroom/id/2684616.
142
Cisco, “Cisco Visual Networking Index (VNI) Global and North American Mobile Data Traffic Forecast Update
(2013-2018),” February 2014 at 28.
143
Jerome Nadel, “Internet of Things Hype Needs a Health Dose of Design Thinking,” Wired, December 19, 2013,
available at http://www.wired.com/2013/12/internet-things-hype-needs-healthy-dose-design-thinking/: “Today’s
45 | P a g e
B. Continued investment and innovation competition
97.
Considerable investments by broadband providers in upgrading wireline and wireless
networks are expected to continue in order to meet this growth in demand. Studies project that,
despite significant capital expenditures by U.S. wireless broadband providers in the past few
years, these investments will need to continue in order to build the networks necessary to meet
consumer demand for wireless broadband.
98.
Investments in wireline networks: Analyst reports project that significant ongoing
investments from cable broadband providers will be necessary over the coming few years in
order to meet the growing demand for broadband. For instance, Comcast’s capital expenditures
are projected to rise from about $6.6 billion in 2013 to $7.9 billion by 2016.144 Time Warner
Cable’s capital expenditures are projected to rise from about $3.2 billion in 2013 to $3.8 billion
in 2016.145 Verizon and AT&T are also expected to continue sizeable infrastructure investments
in their wireline FiOS and U-verse offerings, respectively.146
99.
Investments in Internet backbone infrastructure also have been sizeable and are projected
to continue growing.147 For the U.S. wireline industry as a whole, including the last mile and the
backbone, one recent report projects that infrastructure spending will rise from $38.6 billion in
2013 to $42.9 billion by 2017 (see Figure 12 below).
power, bandwidth and processing infrastructure isn’t quite ready to support 20-30 connected devices, and the greater
infrastructure isn’t quite optimized to quickly and effectively move the data.”
144
Bank of America Merrill Lynch, “Comcast Corp: Throwback to Pay-TV roots,” April 22, 2014 at 2; Bank of
America Merrill Lynch, “Comcast Corp: A ‘positive’ surprise,” January 9, 2014 at 2.
145
Bank of America Merrill Lynch, “Time Warner Cable Inc.: In the Shareholder Interest,” February 17, 2014 at 2.
146
See, e.g., Bank of America Merrill Lynch, "Global Wireless Matrix 1Q14 The Quad Play: How Disruptive?"
April 21, 2014 at 159, 193.
147
U.S. Internet backbone investment has grown from $12.9 billion in 2010 to $20.1 billion in 2013, and is
projected to increase to $24 billion by 2017. (Telecommunications Industry Association, “TIA’s 2014-2017 ICT
Market Review and Forecast,” 2014 at 3-44.)
46 | P a g e
Figure 12: U.S. Total Wireline Broadband Investment
$50
Projection
$45
$42.9
$39.7
$40
$40.6
$38.6
$38.9
2013
2014
$41.5
$37.1
$35
Billions U.S.
$30
$29.4
$29.6
2009
2010
$25
$20
$15
$10
$5
$0
2011
2012
2015
2016
2017
Source: Telecommunications Industry Association, "TIA's 2014-2017 ICT Market Review and Forecast," 2014 at 3-46.
100.
Investments in wireless networks: Increasing demand for wireless broadband will
require increased investment by wireless broadband providers. Studies project capital
expenditures by U.S. wireless broadband providers of around $35 billion a year from 2013-2017,
up from around $25 billion a year for 2010-2012 (see Figure 13 below). The limited availability
of spectrum is a particular challenge for wireless broadband, requiring significant investments
and innovative technological solutions that must be developed by industry participants.148
148
See, e.g., Telecommunications Industry Association, “TIA’s 2014-2017 ICT Market Review and Forecast,” 2014
at 5-15: “[s]pectrum is the key constraint. Spectrum in the 900, 1800, 2100 and 2600 MHz bands is being used for
LTE, and the more recent availability of spectrum in the 700 MHz and 800 MHz bands has helped foster new LTE
rollouts using those bands. Small cell deployments will play a major role in the near-term in alleviating congestion
by offloading traffic to the wired network. Going forward, it is envisioned that spectrum below the 700 MHz band
will be used for a combined broadband and broadcast network. In addition, unlicensed spectrum in the 5 GHz band
could provide additional capacity to offload traffic, with spectrum in the 60 GHz band possibly serving this function
in the future.”
47 | P a g e
Figure 13: U.S. Wireless Broadband Provider Investment
$45
Projection
$40
$34.0
$35
$34.8
$35.0
$35.1
$35.6
2014
2015
2016
2017
$30
Billions U.S.
$25.2
$25
$24.4
$21.1
$20
$20.0
$20.2
$20.4
2008
2009
$24.9
$25.3
$25.7
2010
2011
2012
$16.0
$14.1
$15
$10
$5
$0
2002
2003
2004
2005
2006
2007
2013
Note: 2012 estimated based on doubling investment amount for first half of year.
Sources: Federal Communications Commission 16th Annual Mobile Wireless Competition Report, Table 8; PCIA (Wireless Infrastructure Association),
“Wireless Broadband Infrastructure: A Catalyst for GDP and Job Growth 2013-2017,” Table A; CTIA (Wireless Association), “US Investment Four Times
More in Networks,” available at http://www.ctia.org/resource-library/facts-and-infographics/archive/us-investment-networks.
101.
However, the biggest challenge is that the future of broadband use is highly uncertain.
How consumer usage patterns and demands will evolve, and the business models that will
develop to meet consumer demands are largely unknown. The relatively short history of the
Internet is filled with examples of firms introducing new business models that have changed how
consumers use the Internet. The tremendous innovation that has occurred in the past few years is
expected to continue, and future innovation may dwarf the advances of the past. As I discuss
below, this innovation will require experimentation with new services, business models, pricing
structures, and network management.
V.
Implications of Market Competition for Proposed Regulation
102.
The Commission should give broadband providers broad flexibility to experiment with
and implement novel service offerings and business models. As I discuss below, there are
limited risks and significant benefits to competition and consumers of allowing differentiated
service models, network management, and pricing arrangements with content providers.
48 | P a g e
A. Limited risk that differential service models, network management, and pricing
arrangements would harm consumers and competition
103.
Given the competitiveness of the industry, there is little risk that broadband providers
would adopt differential service models, network management, or pricing that would harm
competition and consumers. The existence of firms with market power, or limited competitive
constraints, is a necessary condition for firms to have the incentive and ability to enter into
vertical business arrangements that harm competition and consumers. But the vast majority of
consumers have the ability to choose among different broadband providers offering high-quality
broadband services. This is particularly true in areas where fiber and DOCSIS 3.0 cable services
compete head-to-head, and also within the highly competitive wireless broadband marketplace.
And, while legacy technologies such as DSL are typically slower than these more advanced
services, they nonetheless compete to some extent for some customers (e.g., customers who may
not put a premium on speed or who are price-sensitive). Moreover, even if some share of
consumers have limited broadband options, the behavior of broadband providers with respect to
those customers is likely to be constrained by competition for other customers who do have
strong competitive alternatives. That is, competition for consumers with high-quality broadband
alternatives also can discipline broadband provider conduct with regard to consumers living in
areas without the same alternatives.
104.
The competitive constraint on broadband providers is buttressed by the fact that
information on provider practices is widely available. Consumers can monitor the practices and
performance of their broadband provider, such as speeds to download particular content, and
switch to a rival provider if they cannot get adequate access to the content they desire.149
Numerous third-party websites and publications provide detailed information to consumers, and
an active online community closely monitors providers’ practices. And, the Commission’s own
transparency rules will further ensure that the information necessary to evaluate the performance
of broadband providers is available to consumers.150 Moreover, because broadband consumers
149
A recent Consumer Reports survey found that 71 percent of consumers “would attempt to switch to a competing
Internet service provider (ISP) if their provider were to try to block, slow down, or charge more for” certain content.
(“71% of U.S. households would switch from providers that attempt to interfere with Internet,” Consumer Reports,
February 18, 2014, available at http://www.consumerreports.org/cro/news/2014/02/71-percent-of-householdswould-switch-if-provider-interferes-with-internet-traffic/index.htm#survey.)
150
Open Internet Order, ¶¶ 53-61.
49 | P a g e
often “multi-home”—i.e., use various different broadband providers (such as a wireline
broadband service at home, a wireline broadband service at work, and one or more wireless
broadband services)—consumers can compare the performance of broadband networks in terms
of the speed and other aspects of the quality of transmission.
105.
The risk of losing customers presents a substantial economic threat to broadband
providers. The provision of broadband access services, both over wireline and wireless
networks, entails significant fixed (and sunk) costs of deploying the network and relatively low
marginal costs of serving existing subscribers. As a result of these fundamental economics of the
broadband industry, the expected revenues from a broadband subscriber during their average
expected lifetime of use (referred to as the “life-time subscriber value,” or LTV) is substantial.151
And, accordingly, broadband providers spend considerable sums in competing to attract new
subscribers (referred to as “subscriber acquisition costs”).
106.
The ability of consumers to switch broadband providers, and the potential for substantial
foregone revenues from subscriber defections, creates significant incentives for broadband
providers to implement business practices that benefit customers. The risk of losing subscribers
also provides a significant competitive constraint on broadband providers in implementing
service models, pricing arrangements, or network management approaches that are harmful to
consumers. The proliferation of consumer review sites, user forums, and blogs intensifies this
competitive constraint, by providing a means for dissatisfied customers to inform and persuade
other consumers, thereby putting pressure on broadband providers to offer the optimal service
possible for their users.
B. Differential network management and pricing arrangements have the potential
to enhance competition and benefit consumers
107.
While there is very limited risk that broadband providers would adopt pricing and
network management approaches that would harm consumers and competition, the benefits of
allowing flexibility in the business models adopted by broadband providers are potentially very
151
For example, a recent study estimated the “Customer Lifetime Value” of a cable subscriber to be between $2,000
and $5,000 for customers that subscribed to a service package which included broadband (e.g., video/data,
video/data/voice). (Robert F. Cruickshank III, “Financial Model for Measuring Cost Savings Driven by Service
Operations that Impact Customer Experience and Voluntary Disconnects,” 34(4) Broadband Journal 66 (2012).)
50 | P a g e
large. Arrangements between broadband providers and content or application providers are
vertical in nature. Vertical arrangements are generally procompetitive, and the economic
literature recognizes various efficiencies that can be achieved through vertical contracts, vertical
restrictions, and other arrangements.152 These efficiencies can benefit consumers in the form of
lower prices, higher quality, greater product variety, and/or enhanced innovation. This is not to
say that vertical arrangements can never be anticompetitive, but such unique instances can be
addressed through ex post enforcement, and do not justify ex ante regulation that imposes a
blanket ban or other substantial restrictions on vertical contractual arrangements between
broadband and content providers.
108.
Differential network management and pricing arrangements have the potential to generate
significant benefits for consumers, competition, and innovation. For instance, commercially
negotiated “two-sided” pricing approaches could benefit consumers, content providers, and
broadband providers alike. Most directly, positive prices to content providers would tend to
lower prices to subscribers.153 This is because a larger subscriber base enhances the value of the
network to content providers, and therefore raises the potential revenues that a broadband
provider might earn from content suppliers if it offers attractive negotiated arrangements.154 The
ability to earn revenues from content providers therefore enhances competition for subscribers,
and creates incentives to lower prices and increase the quality of services offered to subscribers.
Such two-sided pricing arrangements clearly can benefit subscribers. And, the content provider
also can benefit if the negotiated arrangement leads more consumers to use its services. The
possibility of entering commercially negotiated arrangements with content providers also can
provide a source of revenues to invest in providing high-quality broadband services at low
prices, while also providing an incentive to attract more customers to the broadband network.
152
Dennis W. Carlton and Jeffrey M. Perloff, Modern Industrial Organization, Prentice Hall (4th Edition), May 13,
2004 at 414-431.
153
One example of a two-sided pricing arrangement is sponsored or partially sponsored data arrangements where
subscribers are on usage-based pricing plans. For instance, a content provider may enter into an arrangement with a
broadband provider such that subscribers can use the content provider’s services without the traffic counting towards
the subscriber’s usage-based charges.
154
This is referred to as cross-platform network effects, whereby the presence of members of one user group on a
platform increases the value to members of another user group. (See, e.g., Michael L. Katz and Carl Shapiro, “Systems
Competition and Network Effects,” 8 Journal of Economic Perspectives 93 (1994).)
51 | P a g e
109.
This interconnection between pricing on the two sides of a network or platform—in this
case, the content and subscriber sides— has been discussed extensively in the economics
literature on “two-sided” markets.155 Importantly, the economics literature on two-sided markets
shows that there is no economic basis that mandating “free” access to one side of the market (in
this case, content providers) is good for competition, consumer welfare, or innovation.156
110.
Although, to date, the competitive equilibrium seems generally to be one where last mile
broadband providers have not directly entered into pricing arrangements with content and
application providers, industry developments may make two-sided pricing arrangements
beneficial.157 Diverse and innovative pricing and network management arrangements are likely
to benefit consumers in the future, especially given the significant diversity of applications, uses,
and consumer preferences.
VI.
Conclusions
111.
Given the competitiveness of the broadband industry, any “Internet Openness” rules
adopted by the Commission should give broadband providers broad flexibility to implement new
and novel service offerings and business models, including differentiated network management
and pricing arrangements with content providers that have the potential to provide valuable
alternatives to consumers. Ex-ante regulation that limits flexibility in business arrangements
would be especially inefficient given the nascent and rapidly changing broadband environment,
155
See, e.g., Benjamin Klein, Andres V. Lerner, Kevin M. Murphy & Lacey L. Plache, “Competition In Two-Sided
Markets: The Antitrust Economics Of Payment Card Interchange Fees,” 73 Antitrust Law Journal 571 (2006);
Richard Schmalensee, “Interchange Fees: A Review of the Literature,” 1 Payment Cards Economic Review 25
(Winter 2003); Jean-Charles Rochet & Jean Tirole, “Cooperation Among Competitors: Some Economics of
Payment Card Associations,” 33 RAND Journal of Economics 549 (Winter 2002); Julian Wright, “The Determinants
of Optimal Interchange Fees in Payment Systems,” 52 Journal of Industrial Economics 1 (March 2004).
156
See, e.g., Benjamin Klein, Andres V. Lerner, Kevin M. Murphy & Lacey L. Plache, “Competition In Two-Sided
Markets: The Antitrust Economics of Payment Card Interchange Fees,” 73 Antitrust Law Journal 571 (2006);
Richard Schmalensee, “Interchange Fees: A Review of the Literature,” 1 Payment Cards Economic Review 25
(Winter 2003); Jean-Charles Rochet & Jean Tirole, “Cooperation Among Competitors: Some Economics of
Payment Card Associations,” 33 RAND Journal of Economics 549 (Winter 2002).
157
However, there are instances in which content providers “sponsor” their customers’ data usage. For example,
Amazon’s Kindle Paperwhite reader offers wireless broadband connectivity with no monthly fees or contracts for
the user. Moreover, as discussed, it is common for content providers to pay to enable faster and more reliable
content delivery, including through CDNs, which suggests that there may be demand by content providers for
prioritization arrangements.
52 | P a g e
and the significant uncertainty regarding broadband industry developments in the coming years.
There is significant uncertainty regarding how consumer usage patterns will evolve in light of the
ever-changing array of online applications and services that are available, the types of broadband
business models that may emerge, and what will be the optimal network management and
pricing arrangements. As a result, there is no reason to believe that restricting business
arrangements implemented by broadband providers, without a clear showing that such
arrangements inhibit competition, would enhance consumer welfare. More likely, ex ante
regulation that restricts the ability of providers to experiment with and implement legitimate
business arrangements would impose considerable costs, distort competition and market
outcomes, and reduce investment incentives, to the detriment of consumers.
I declare under penalty of perjury that the foregoing is true and correct.
____________________________
Andres V. Lerner, Ph.D.
July 15, 2014
53 | P a g e
Appendix A: Qualifications
112.
I am an Executive Vice President at Compass Lexecon. I received my bachelor’s degree
in Economics from the University of California at Berkeley and my Master’s and Ph.D. degrees
in Economics at the University of California at Los Angeles (UCLA). My areas of economic
expertise are antitrust, industrial organization, regulation, and econometrics. I have applied my
expertise to a wide variety of industries, including wireless telecommunications, broadband
services, mobile devices, cable and satellite television, online search, e-commerce, online
advertising, payment cards, pharmaceuticals, semiconductors, motion picture, video games,
automotive, and airlines. I have presented economic testimony and submitted declarations in
regulatory proceedings in several forums, including the U.S. Federal Trade Commission, the
Antitrust Division of the U.S. Department of Justice, and the U.S. Federal Communications
Commission.
113.
I have been named one of the foremost competition economists in The International
Who’s Who of Competition Economists for 2014. I have published scholarly articles in leading
economic and legal journals, including the American Economic Review, the Antitrust Law
Journal, and the Antitrust Bulletin and have co-edited a collection of seminal articles in antitrust
economics. I have taught Economics as a Visiting Professor at the University of Southern
California Marshall School of Business. I am a member of the American Economics Association
and the American Bar Association.
54 | P a g e
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement