FTC NN6
- See FTC Workshop Transcript for index
Quality of Service, Tiering, and Charging Fees for Prioritized Delivery
This panel addresses issues relating to data prioritization by network operators, including quality-of-service requirements, tiering, and charging fees for prioritized delivery. Among the topics for this panel are: Current pricing and data exchange arrangements among the relevant parties; the likely effect on competition in the markets for content/applications and broadband access; the likely effect on innovation at the edges of the Internet and at the network core; the two-sided nature of the broadband market; network capacity constraints and the need for prioritization; and what the end result of prioritization might be.
>>MAUREEN OHLHAUSEN
If everyone can take your seats so we can finish up and get you out of here on
time.
>>WILLIAM BLUMENTHAL
Good afternoon, everyone, I'm the agency's General Council. I would like to
welcome you to the last panel of the afternoon. We'll go until about 5:15 today.
The good news, for those who haven't heard, the Federal government just closed.
They closed at about 2:00 P.M.. By 5:15, the congestion problem on the road
should be largely solved. At the same time, the roads are pretty slick and I
suspect this experience will give new meaning to the term jitter. In all seriousness,
the salt crews are out there, so the extra hour or two may actually work to your
benefit. For this panel we'll be focusing on prioritization, and the charter of the
panel is up on the screen. I'll just go through it quickly what we'll be talking about
which is quality of service, hearing, the prospect of charging fees for prioritized
delivery. And there are a whole lot of things that that in turn maps into. The topic
in many ways, the title is different from the title of the last panel, but prioritization
isn't all that different conceptually from discrimination, and I suspect that the talk is
going to focus on many of the same issues. We'll try to be a little bit more of sort
of a technical adviser as opposed to a policy advisers we'll see what emerges.
Those would are Shakespeare fans know that Richard III is playing a few blocks
down the street. If you run that two or three or four times the way we're running
similar types of themes, there is somewhat different variation in how it presents.
So we're going to be going through again some familiar issues but with a slightly
different twist. Each of the panelists is going to speak for about 10 minutes with a
kickoff set of comments. If they are a little long I'm not going to give them the
cane.
>>MALE SPEAKER
You should not tell us that.
>>FEMALE SPEAKER
Don't worry, we will.
>>WILLIAM BLUMENTHAL
I'll tell you in all seriousness, order of presentations was chosen by lot. There is a
story behind that but I'm not going to share it right now. First-off, Alan Davidson,
who is Washington Policy Counsel for Google, a company that I'm sure is known
to all of you. Those who are in the industry, would know Level 3 as well but for
those who are not regulars in the industry, our second speaker John Ryan is
Senior Vice President and Assistant General Counsel for Level 3 and Level 3, for
those who are not in this stuff day-to-day, but anybody who is in it day-to-day
would know Level 3 is one of the six tier one backbone suppliers in the U.S.
basically. Something that emerged from the old (inaudible) way back when. But
fair to say, John, the wholesale side of things?
>>JOHN RYAN
Arguably now in the retail side of things as of last year.
>>WILLIAM BLUMENTHAL
Best known probably as a bock bone supplier. The third speaker will be Walter
McCormick, who is the President and CEO of the United States Telecom
Association, which is a trade association representing the term that's used as the
converged telecommunications industry. I think probably it's fair to say that most
often associated with the big telecom, is it fair to say?
>>WALTER MCCORMICK
Well, our 800 small members think that. We have two large members, AT&T and
Verizon. We have 800 smaller members.
>>WILLIAM BLUMENTHAL
Fair enough. Fourth up is Marius Schwartz, a Professor of Economics at
Georgetown, he has a lot of experience in the telecom industry including years
ago at DOJ antitrust division if I recall correctly. The fifth speaker will be Barbara
Tulipane, who is the President and CEO of the Electronic Retailing Association,
which is a 500-member trade association based here in D.C. that represents a
large portion of the electronic retailing industry. So with that, actually, one more
reminder. We're taking questions in the manner that was specified earlier in the
morning. I think there are question cards. If you do have a question, hold up the
card and someone from the FTC staff will come on down. Probably we'll be going
50 minutes, 60 minutes of straight through talk, but with Q and A after that. So
with that, let me turn it over to our first speaker, Alan Davidson.
>>ALAN DAVIDSON
Thank you. I would like to start by saying thank you to the FTC and its staff for
organizing another of the very thoughtful workshops that they have become known
for, that are so valuable in exploring these complex and I guess apparently
occasional abysmal discussions here in Washington, so we really appreciate that.
I'm Alan Davidson, Senior Policy Counsel for Google in Washington. I wanted to
quickly touch on three things that I wanted to cover briefly. First of all, why we
believe net neutrality is so important for the health of a competitive internet, briefly.
Secondly, the problems and risks that we see with certain types of last mile
router-based prioritization, which I think are the core of the issue that we're here to
discuss. And the third thing is, to just talk briefly about some of the myths that we
see surrounding prioritization and the net neutrality debate. So first, net neutrality
is critical for the health of a competitive internet. We've heard a lot about this
already today. The internet has created one of the most innovative and
competitive markets in history. Services that we could never have imagined even
a few years ago, now drive economic growth, democratic discourse around the
world and the free exchange of ideas. Much has been said about this. Success
has been a father's -- many people acknowledge that this innovation has been
made possible by the architecture of the internet. The open architecture of the
internet. On the internet consumers choose what they are going to see and do.
What services they have access to. What content they are going to access.
There are no gatekeepers to tell them what they can see and do online. And that
principle was a conscious design choice made by the founders of the internet. It
has enabled innovation at all layers of the network. Not just at the edge but, in
fact, in the network itself. And we value that innovation. It's that principle that
actually enabled companies like Google to rise. Google didn't exist eight years
ago. We're a second grader. And we now help nearly 500 million people, users
around the world every month, find information and reach services online. Billions
of searches done here in the United States alone each month. As our founders
have said, two graduate students in a dorm room with a good idea, would not have
been able to create this service if the first thing that they had to do was to higher
hire an army of lawyers and reach carriage agreements with providers all around
the world. So we're very eager to preserve the innovation and openness of the
internet that has allowed companies like Google to develop. I sense that there is
actually a lot of agreement around that idea. We're here because there has been
a sea change that puts that openness in jeopardy, all the things that people have
been talking about, we'll continue to explore. The situation in terms of
competitiveness in the last miles. The change in the long standing rules that have
governed the openness of the on-ramps to the internet. And the stated intentions
of some of the last-mile providers in terms of what they hope to do and achieve in
this environment. And because of that, we think it's very important that we're
having these discussions, that we have a dedication to try to preserve network
neutrality and I think that's why we've seen such an outpouring of small
businesses, consumers, public interest groups from the right, the left, the Christian
Coalition, the AARP, Consumers Union, a million internet users who signed a
petition last year, a save the internet campaign to preserve the openness of the
internet so this is something that's obviously of great concern and should be of
great concern to internet users. My second point is that prioritization in the last
mile creates real concerns. Particularly we're concerned that prioritization through
router-based discrimination in the last mile degrades competing services and
creates incentives to regulate some of those competing services to a slow lane.
So it's this particular set of prioritization approaches that we're concerned about.
What we're worried about is in that context, the power to prioritize in the last mile
effectively becomes the power to control the applications and content that
customers can effectively use. Imagine, for example that a last mile provider with
market power might be able to use prioritization to, for example, relegate a
competing voice over IP provider to an upper or lower quality slow lane. It might
provide a competing video provider -- competing video service from accessing a
higher tier of priorities necessary to provide good service. Preference its own
services instead. Not all network management is anti-competitive prioritization.
There are a lot of things I think many of us agree that are not problematic in this
context. So charging end users, whether it's businesses or consumers more for
more bandwidth, not a problem here. Providing caching services, like Acamy (ph)
does, not a great concern. Creating dedicated IP TV channels for television
services, none of us have argued that last mile provider shouldn't be able to do
that. We welcome that kind of competition for the existing cable television
networks. Stopping denial of service attacks, not a problem. We think those are
reasonable network managements that should not be precluded by network net
neutrality. The problem is with a very small set of prioritization activities in the last
mile, the ones that give carriers the incentive to degrade competing traffic and pick
winners and losers in the last mile so what we're worried about is, prioritizing some
traffic at the router level, in the last mile at the expense of other traffic. That's one
thing. We're worried about blocking traffic in order to preference other traffic.
We're worried about degrading traffic, the same way that Rogers cable in Canada
degraded network video traffic there. We're concerned about creating a fast lane
tier of traffic that's susceptible of exclusive dealings, so things that provide an
incentive for there to be a slow lane. That's really, you know, I think the core of the
concern, is that the only way that you can have a fast lane that you can charge for,
that is useful, is if there are also slow lanes that are less useful and less attractive.
And so prioritization that provides an incentive to create slow lanes so that you can
charge people for the fast lanes, is something that we think is problematic. Some
of the biggest impacts of that kind of prioritization, probably the first and foremost
is that it puts new entrants at a major disadvantage. Only those with the ability to
pay will be able to benefit from this prioritization. And so we're quite concerned
that the next Google will have a very difficult time being able to get access to these
faster lanes. As I've said, that could give a great deal of control over the future
services that consumers have access to to the last mile providers. I also note that
it's not clear to us exactly how this will work. You can't control priority end-to-end,
right? So last mile providers who want to give some sort of priority service, you
know, only have control over their own network. It's not obvious to us, how you
can I.D., how you identify the traffic in order to segregate it, that you're going to
give priority to. And how do you do the segregation without degrading other
traffic? Very quickly, a couple of the myths surrounding prioritization. In most
cases, prioritization is a solution in search of a problem. It's not clear that there is
a compelling need for last mile router-based prioritization. VOIP is a great
example in excess of 100 million happy Skype users who are getting excellent
voice service over the internet without a prioritization regime in place. A lot of
VOIP providers are providing that service without -- over narrow band connections
so this notion, for example. Already mentioned a couple of times today that you
need prioritization to make services like voice work, is just simply not true. In most
cases the best way to deal with any concerns about prioritization is provide better
broadband, higher bandwidth offerings to consumers and that will be the way to
deal with prioritization. The other thing I would do is note that prioritization is not
needed to fund network rollout. Another argument is made is hey if you don't let
people do prioritization they will never be able to fund all of the investment in the
networks. There are billions of dollars being spent by consumers and businesses
to access the internet. There are billions more in special access fees being
shared by broadband providers there will be new IP TV offerings, there will be
caching services. All of those are excellent things and should be -- we welcome
the chance for broadband providers to have these great incentives to invest. It is a
very small set of prioritization activities that we worry about and also, a tiny part of
the full pie of income that's going to -- that will pay for this broadband rollout. We
really actually questioned how valuable these prioritizations services are going to
be ultimately to the providers, and so we think, you know, I'll just summarize by
saying, first of all, we welcome the FTC's involvement here. There has been a
number of ideas put forward about the potential role of the FTC investigating
complaints, requiring disclosure, which we think is a very, a very welcomed idea.
And won't be enough to protect consumers but is a very good starting place, and
the kinds of approaches that have been put forward, for example, by the House
Judiciary Committee in their bill last year. All of those are good places for us to
look in terms of ways to deal with the real concerns here. There is a good deal of
agreement act the fact that more broadband deployment, open broadband
deployment is good for our country, for consumers. A thriving internet
marketplace is good for consumers and for the industry. Providing incentives to
deploy broadband is critical. We're in a symbionic relationship here. The
broadband providers -- internet industry needs more broadband deployment to
get our services out there. We welcome it. At the same time, we're providing the
services and content that drives the demand for that new broadband. We need
each other and need to find ways to work through This hopefully we'll find that
there is a very small set of things that we really need to work on, and we look
forward to working together with all parts of the industry and the commission and
the consumer groups to find ways to get America the open broadband that it
needs. Thanks.
>>WILLIAM BLUMENTHAL
Thank you. John Ryan, agree or disagree?
>>JOHN RYAN
Agree in some respects with the overall principle but I think we do have a slightly
different take on what has to happen. I'll start with a confession, which is
generally, you people scare me. There is a fair amount of intelligence in this room
that I can't hope to match as is evident from the questions that I got even during
the break. I want to touch on three areas. I'll start by giving you some perspective
on our view of the debate. A little bit of open disclosure on what we think the
solution is to this risk. Then I want to touch on the current network reality. We
operate a very large IP network. By some measures, the biggest in the world. I
think we can share with you what we're seeing happening in the network or
potentially happening in the network from a prioritization perspective. Finally, I
would like to discuss some possible future incentives to avoid what I would call
some anti-competitive prioritization by access network operators. It seems to me
that we have two competing policy objectives here. One is, we want to preserve
an open and dynamic and ever-changing internet experience for all of the
subscribers. The second is we need to encourage the continue migration to
broadband services, and frankly encourage and increase the speed and the
performance of the broadband services that are delivered to subscribers.
Broadband in the hands of consumers is an extremely powerful tool. Sometimes
contains powers that even the networks that deployed it didn't understand. I'll
demonstrate by giving you an absurd or ridiculous hypothetical. Let's assume it's
five years from now and broadband over power line has become perfected. And
it's being delivered in the marketplace. Let's assume at the same time, some
innovative applications designer figures out a way to create a marketplace for the
purchase of electricity. So that you can buy electricity from competing electrical
providers over your BPL connection. Now, it doesn't take a Ph.D. in economics to
figure out that that's going to cause concern to the BPL providers. They now have
a potential conflict between these two principles. If we permit blocking or
degradation of those electrical purchase bits, the subscriber's use of the internet is
potentially impacted. On the other hand, if we prohibit blocking, the companies
with the ability to deploy broadband may not do so because they are going to
cannibalize their core revenue. This absurd and ridiculous hypothetical already
happened. Companies in the late 1990s and at the turn of the century were
deploying broadband over DSL only to find out two years later, bingo, Vonage.
And those same companies that were deploying DSL suddenly realized, they are
putting into the hands of their subscribers a tool to make them irrelevant. These
are two very important objectives from a policy perspective, we need to keep our
eye on how to satisfy both those objectives as we look at potential legal solutions
or policy solutions to these issues. We're not an advocate of a network neutrality
mandate. Given the choice between regulation to solve a problem and allowing
the marketplace to solve the problem, we're fans of the market. I was listening
earlier when somebody said that what we're really talking about is looking at
regulating or somehow addressing IP interconnection. Our company has
experience on both regulated interconnection with a public switch to telephone
network and nonregulated interconnection on the IP side. I have to tell you,
regulated interconnection stinks. It may have gotten us where we are on the
public switch telephone network, but it is horribly inefficient. And if you look at the
innovation that's been done on the PSTN over the last 100 years, namely zero,
you'll realize that one of the reasons for that is regulated PSTN interconnections.
When you're sitting down and talking to those folks, they do what's in the regs and
that's it. I have five full-time attorneys and a $2 million outside counsel budget to
handle PSTN interconnections. I have half an attorney and zero outside counsel
budget to handle IP interconnections. So if you want to regulate IP
interconnection, understand the wet blanket that you're potentially throwing on
what appears to be working right now. Now that we've covered some of the
background, I'd like to discuss what prioritization is happening. The OECD I think
has a paper circulation that actually has a very good discussion of packet
prioritizations. I'll borrow a little bit from that. I will tell you that I will correct a little
bit in terms of erroneous terms that I see in the OECD discussion. They separate
prioritization into three different areas. What they call best efforts prioritization,
what they call needs-based prioritization, and then what they call active
prioritization, which I'll call source or type prioritization. That's what Alan was
referring to as the third type of potential prioritization. First, let me be clear. IP
networks do prioritize. They have from the beginning of time. The prioritizations
that they had in the network at its inception was basically a first inline prioritization.
First in, first out. So it's prioritization based on time and time alone. That appeals,
I think, to our fundamental sense of fairness as a society, although I now am a
frequent flyer, so I get to violate that but its like the cuts in line that the see at the
airport doesn't seem fair. Gee, that guy got to cut in line. Doesn't seem fair. So
this notion, first in, first out, is the fairest and therefore should be the mandated the
approach to prioritization, we think is a little bit too broad. The second is
needs-based prioritization and this is happening in the network right now.
Need-based prioritization is a situation where the customer or the user of the
network identifies the packets that require delivery quickly. So there are certain
time sensitive applications that are already running over the internet. Video over
IP is one example, live streaming video. Our customers don't like delay and they
don't like jitter. And there are ways to reduce delay and jitter by giving packets
prioritization. Largely, this occurs in the network right now through the purchase of
what are called IVPVNs, or virtual private networks that operate on an IP basis.
It's the functional equivalent of a dedicated circuit, but it operates IP. And then
within that dedicated circuit frankly, those customers then will prioritize some of
their packets over others. So that's needs-based prioritization. Then there are
source-based or type prioritization. This is reading a header in a packet, every
packet has a header that tells you what port it's destined for, where it's coming
from and what sort of packet it is. The routers now have advanced to the point
where you can read that header and then you can tell the router to treat that
packet differently based on what's in the header. Priority frankly matters most in a
constrained capacity environment. We do prioritize on our network at times, but
our backbone network runs on multiples of 10 DBD(ph). So we've got 2 1/2
networks effectively deployed where one would be sufficient in order to account for
bursts in traffic, in order to account for a portion of the network being pulled down
and those routers having to rout traffic over a different physical infrastructure. So
on our network I can tell you we'll put two packets in at one end in Los Angeles
and say, let's give one priority and let's not give the other one priority and see what
happens when they arrive in New York City, and the answer is, you can't tell the
difference between the two. At least not right now on the backbone. Now, on the
access networks, priority does matter. Because those networks are not built up to
10 DBD (ph), and frankly those networks are not built out to even handle, if you
took all of the subscribers who sit behind a particular central office, for example,
and you said, everybody is purchasing four megs of traffic, so cumulatively we
have 100 people purchasing four megs of traffic, there should be 400 megs of
capacity into and out of it, that capacity is not there. It's theoretically there for all of
those users but the truth is, if everybody is using at capacity, the network can't
handle it. So that's where prioritization matters, is in those last mile networks.
What's the risk -- let me back up. I will say I'm not willing to concede that all forms
or source of prioritization are anti-competitive. Likewise, I'm also not willing to
concede that first in, first out routing, is always pro-competitive. So let me give
you two examples of why. First, let's assume there is a new application that
somebody has developed that requires a little bit better network performance on
the edge of the network in order to operate. That better network performance can
be achieved by prioritizing the packets associated with that application. I'm not
sure that I'm willing to concede that it's anti-competitive for Verizon to say, gees, I
would really like to deliver that to my subscribers, so I'm going to prioritize it. Now,
if they are able to give away priority, I'm also not sure that I see the difference,
meaningful different between giving it away and selling it because if you allow
them to give it away, eventually, they will get value for it in some way, shape, or
form. At the same time, first in, first out, may not be pro-competitive in every
instance. There are circumstances where first in, first out theoretically, I think it's
only theory right now, as Alan indicated, where first in, first out, doesn't create time
sensitive packets sufficiently well, and could be viewed as anti-competitive.
>>MALE SPEAKER
Thank you.
>>MALE SPEAKER
Bill, thank you. I want to join with Alan and John in thanking you and the
Chairman and the members of the commission for organizing this event today.
This is a very very important public policy debate. This is the information century.
We're an information-based economy. Some people refer to this as a net
neutrality debate. But what we're really talking about today is regulation of the
internet. It's about whether the government should create and establish rules that
would dictate what kinds of services can and cannot be offered and how
broadband networks can and cannot be engineered and operated. What do we all
really want to see next? What we want to see next is a better, faster, and a more
robust internet. It's getting easier and easier to imagine a world in which you can
sit at home and talk to your doctor as he or she looks over your blood pressure or
heart rate in real-time. Or work collaboratively with a team from the office without
ever leaving your home and spending an hour in traffic, a world where if your child
needs tutoring, or if you have a sick parent, they can get the help that they need,
help that is far more sophisticated in terms of the communications technology than
a plain old telephone call. We're confident that a broadband future means a
variety of consumer applications that have enormous societal benefits. Broadband
can make it possible, but it will take policies that encourage investment. It will take
policies that understand how the internet works, and reflect the importance of
network management. Quality of service. And prioritization. A better internet
doesn't simply come by adding capacity. Like road networks, rail networks,
electrical networks, and traditional telephone networks, the advanced networks
that comprise the internet cannot function efficiently and cost effectively without
management. No network has ever been built without regard to prioritization of
traffic, peak loads and capacity management. Indeed, as John said, traffic is
managed on the internet today, network management reduces spam. It controls
viruses. It enables a host of privacy and security measures which protect
consumers. It allows us to manage jitter and latency making possible phone calls
over the internet that we can actually understand. It makes possible video
streaming. With regard to prioritization in particular, no doubt that we can all agree
that certain services are objectively more important. A communication about your
health, for example, is clearly more important than how quickly your kid can
download a video featuring the antics of someone's pet hamster. Those who say
outlaw prioritization, prohibit discrimination among bids, require that all packets be
treated the same and travel at the same speed, which is exactly what legislation
introduced on Capitol Hill right now would require, would prohibit a wide array of
practices that currently increase the value of the internet for consumers. And they
ignore the need to address the capacity issues that present real challenges going
forward. One recent report noted that if YouTube alone goes high def, that would
double capacity needs of the entire internet. The data involved in one hour of
video can equal the total in one year's worth of emails. I'm joined on the panel
today by Alan. Alan's company and the company I usually don't usually sing
KumBaYah in a circle but this past week Google's Chief of TV Technology
expressed concern that the capacity being required by new internet video
services, file swapping and downloads, may overwhelm existing internet offerings
and degrade consumers' overall quality of service.
>>MALE SPEAKER
I don't think that's what he said, so I look forward to having that conversation in the
Q & A. I want to set the record straight that's really not what he said.
>>MALE SPEAKER
It seems like the quotation marks around his comments were taken out of context.
>>MALE SPEAKER
I think they were.
>>WALTER MCCORMICK
These remarks, these concerns about capacity are a welcomed acknowledgment.
If so, if Google has any concerns about capacity. Concerns about capacity, I
think, are a welcomed acknowledgment. The consumer interest in a better
internet come from policies that limit innovation to edge services. We need
investment and innovation and intelligence in the network itself. And the freedom
to engage in network management. So it's important that any national policy
regarding the internet do some prioritization of its own focusing first and foremost
on the consumer. Consumers deserve to have a constantly improving internet
which requires investment and network management. Consumers deserve to
have an internet where they can access any lawful website, where their access is
not blocked, impaired or degraded. Consumers should be able to run any legal
application, consumers should be able to attach any lawful device. As service
providers we have made these commitments. Initially, we made them because
they represent good business practices. But today, they carry with them an FCC
mandate. It one that the Chairman of the FCC has made clear that he has both
the authority and the will to enforce. Consumers can go anywhere they want on
the internet today. They will continue to be able to do so tomorrow. There is no
problem that requires regulation. Perhaps most importantly, the market is
competitive. Speaker after speaker after speaker today has said that, where there
is market power, where consumers lack a choice, there may be a problem. If
there is market power or consumer's lack of choice. The FTC is an agency of
expertise. There is a rigor and a dispassion and an intellectual discipline to the
FTC's approach to competition policy and antitrust analysis. It is an analysis that
exams trade practices without regard to the technology in question but rather with
regard to the characteristics of the market and the behavior in question. So let's
examine the market. First, is it competitive? Do consumers have choices for last
mile access? They do. Today, there are more than 1,200 broadband service
providers in our nation. You can obtain high-speed internet access from your
telecom company, from your cable company, your wireless company, your satellite
company, in coffee shops and on airports and college campuses, and in many
municipalities you can access the internet via WiFi hot spots. Electric utilities are
beginning to invest in delivering broadband over power lines. In fact, analysts
expect 2.5 million Americans to get their high-speed internet in this way within four
years. That's a six-fold increase. Second, do telecom companies have a
dominant share of the broadband market? The last mile market? No, we do not.
The latest FCC data shows that the DSL's market share is at about 36%. Cable's
market share is at about 44%. And we see a rapidly growing other category of
about 20% led by wireless broadband that saw a 58% increase in the first half of
last year. So neither the entire DSL industry nor the entire cable industry has a
market share that rivals one single company's control of the internet search
market. Google is the gateway through which the vast majority of internet
searches go today. Some estimate as high as 70%. Third, do telecom companies
have market power? Defined as the power to control price? Clearly not.
Broadband prices are coming down with entry-level offerings as low as $15 a
month. Speeds are going up and these are signs of a healthy, competitive market.
Google and others are readying plans to offer broadband service themselves for
the imminently attractive price of free. Finally, as, is the market contestable?
Absolutely. With the availability of unlicensed spectrum the rising tide of municipal
WiFi and rapidly expanding BPL options this market is open to anyone who is
willing to invest. In fact, if you think of all the different companies and all the
markets that play a defining role in the ability of a consumer to access and
navigate the internet, the broadband access market is among the most competitive
pieces of the puzzle. Just consider this. In order to access the internet you need
a computer with a chip. There are two companies, Intel, and AMD, who share a
60/40 split of the chip market. That's two companies, not two industries. You
need an operating system. Microsoft and Apple are your basic choices. You need
a browser. Microsoft's share of that, that market is 85%. Internet networks
depend on routers to direct traffic. You're back to two big players, Cisco or
Juniper. And finally, you need a search engine. Here, as I mentioned, Google
controls a share estimated by some to be as high as 70% and climbing. One
company. My whole industry has nowhere near the market share in internet
access that Google has in internet search. So from the standpoint of the FTC's
jurisdiction, there is competition. Consumers have choices. There is no
dominance in the last mile. The market is contestable to anyone willing to invest
and consumers are experiencing no problems. Therefore, we say why would
anyone now start asking for government regulation of the internet? Let consumers
continue to drive the market and they will reap the greatest benefits from the next
generation of broadband innovation. Will, thanks.
>>WILLIAM BLUMENTHAL
Walter, thank you. Alan, I know you'll want to rejoin, but -- we need to hear from
our last two speakers.
>>MALE SPEAKER
You have something to say?
>>WILLIAM BLUMENTHAL
Let me turn the floor over to Professor Schwartz..
>>MARIUS SCHWARTZ
I'll stand. It's the professor in me. If I try to do something differently, it won't work.
I was worried that you might dismiss my remarks as being too hands-off but now I
can be the reasonable man. (laughter) Thank you. Let me agree with many of
the speakers today who said we should try to get past the labels and get down to
the issues. Exhibit one here is the term net neutrality. The traditional model of the
internet where traffic is treated uniformly or first come first serve, priority-based, is
not neutral. If different applications require different network performance then
uniform treatment is not neutral. Point one. So the question is, what kind of
departure from that model are sensible? To date, the debate in the U.S is focused
over what departures should be allowed by residential broadband network
providers, access network providers? On the perception by some that that's
where there is a fair bit of market power and more so than in other segments.
That claim is being contested and I'm sure we'll have more of it. But I don't want to
get into that question of just how much market power there is there because that's
the subject for tomorrow's panel. Let me just make one remark on that, which is,
we're not in the monopoly model. That's for sure. So there are two strong
platforms -- DSL and cable and economics tells us that even a "dualopoly" can be
quite different from a monopoly. Behavior in a duopoly can range from perfect
collusion on one hand to quite competitive on the other. So we're not, you don't
want to presume it's a monopoly firstly. Secondly, it's not to blockage a dualopoly.
There are prospects, how strong I don't know but it's certainly not blockaded. Let's
keep that thought in mind as we move forward. So broadband providers seek
discretion to prioritize, they say, in order to get better network performance.
Prioritize and price based on priority. And in order to finance investments. Okay.
Net neutrality opponents say, wait a minute, you guys have market power. If we
give you unfettered discretion, you can abuse this discretion. And what constitutes
abuse? So we agree you need some discretion to control viruses, spam, et
cetera, but we're worried if we leave it completely unfettered you can abuse it.
What's abuse? Largely the concerns have been about discriminatory treatment of
content and application. As opposed to quality for consumers. I'll focus on the
content and application. To sharpen the analysis, I think it's helpful to distinguish
between two classes of concerns. One is, foreclosure or leverage that was
discussed in the previous panel. The broadband provider reduces competition in
adjacent markets for applications of content, by either giving preferential treatment
to its own affiliates or to favored partners. That's foreclosure. The second
concern is what I'm going to call value-based pricing. Network owners will tailor
the pricing that they charge for transmission services, access services, to how
much the consumers or the suppliers of the applications are willing to pay. Okay.
And they can do that according to net neutrality proponents by threatening, unless
you pay your quality gets degraded. That's the risk of allowing discretion on
quality or any kind of the prioritization, because it could be misused as a lever to
extract payments under this story. That's the so-called tax on applications harm.
And the result is potentially also to harm consumers. Potentially, not necessarily.
So foreclosure has been addressed. I'll just make two quick remarks on that. First
of all, while there can be gains to a broadband access provider from foreclosing,
there can also be losses. Joe Farrell, I thought, presented a fairly nuance
discussion of those tradeoffs. The potential losses come from the fact because
the applications are complementary service to the broadband access, if you
degrade that supply, by freezing out lower cost providers of applications, or
reducing the variety, that will reduce how much consumers are willing to pay for
the access service in the first place. It's not a slam-dunk but it does say there's a
trade-off there. Many of the examples that have been put forth illustrating the
"obvious incentives to foreclose" are drawn from a different era. From a regulative
monopoly era where there are strong incentives to foreclose and those examples
cannot be directly transplanted into this setting. If we have time I can flush that
out. The second point I would make is if foreclosure does rise to a level of a
serious competitive problem, the right response is to address it at the time on a
case-by-case basis. At least that's my view. Stricter remedies like equal access
rules or prohibition against vertical integration as a pre-emptive measure are
things we typically reserve only for a regulate the monopoly regime, as done with
the AT&T's historically in the line of business restrictions. Let me now turn to
value-based pricing. This concern can be quite independent of foreclosure. So if
the network operator simply says, I'm going to charge all applications of voice X
dollars for the right to use the pipes and I'm going to charge video providers Y
dollars, okay, I'm not going to say foreclosing either voice or video, I'm just "taxing"
their services. Now, let's subdivide that into two pieces. The first is, should the
broadband provider be allowed to treat and price traffic differently depending on
the underlying application? The second question is, should the broadband
provider be allowed to "charge applications provider" as opposed to charging only
his consumers? That's the two-sided market question that we've heard of. All
right. So let me take the first one. The differential network treatment in pricing.
Jon Peha this morning gave I thought very nuance discussion of the issue, which
is, differential treatment can be good, can be bad. You can't easily fit and hold it.
The claim that maybe we don't need any prioritization, that just bigger dumb
networks that may be true in a particular context, it may be true that in the
backbone today there is excess capacity. I don't think you want to put forth as a
general design principle. Economically it doesn't make sense that the solution is
always to build more. That will involve carrying a lot of excess capacity which will
be expensive. Tools like prioritization will help you to reduce the amount of
capacity. You can size for maximum congestion, for maximum traffic use. You
instead cope with congestion through prioritization and other tools. In doing that it
makes sense to use the price system as a signal of which things merit priority.
Okay. So -- I think that the most nuance network neutrality advocates would agree
that just building bigger and dumber pipes is probably not the universal answer.
Now, what they say, though, okay, let's try the net neutrality opponents, let's try to
preserve the good aspects of traffic management and prioritization while keeping
out the bad. So two proposals I've heard over time. One is called consumer
tiering but not application tiering, and the other one is allow application tiering but
not for pay. Unpaid application prioritization. Let's take these in turn. Consumer
tiering, the idea, as I understand it, today, we can get different quality of service
but that's largely confined to the size of my connection. You would think down the
road that superior quality may go deeper into the network. I'm not an engineer,
but I can imagine that happening. And the idea is, let's allow that. If you want to
buy higher quality of service for all of your traffic, fine, but I'm not going to allow
different prices and different qualities tailored to the particular application. Well,
the problem with that, consumers may not require a uniformly higher quality for all
of their applications. They may require and be willing, two minutes -- too biblical
minutes. (Laughter) That's a problem with that. Okay. The unpaid application
tiering. What's the logic behind that? The logic, let's allow the network operator to
prioritize, because then he'll make a judgment if voice really needs it, let's prioritize
that above other stuff but we won't let him charge for it because if he has the
ability to charge he can use the threat of withholding quality and degrading quality
as a vehicle to stake out payments from the suppliers. The problem with the
solution is who decides which things deserve and which things don't deserve
priority. The consumers and the application suppliers really are in the best
position to decide that. That gets revealed by their willingness to pay for it.
Moreover, the requirements of priority can differ even for a particular application
across different suppliers. If I'm supplying an application that's high quality, high
priced, and another guy is supplying a low quality, low priced version, I may be
willing to pay for a priority but the other guy may not. The push back is using it as
a price discrimination device. All I'm going to say on that, yes, price discrimination
is not always good, not always bad. Awfully hard to tell it apart and I'm skeptical
that you can do it in a way that doesn't throw out the baby with the bathwater. Let
me take two minutes, I promise, just on the last second and last point. Which is,
should we -- should broadband be allowed to charge application providers like
Google, say. Well, the theory on two-sided economics and two-sided market
approach as a promise follows. It says the broadband provider is an intermediary.
He needs to get both application providers to use the platform and the residential
consumers. If he overdoes it and charges too much on the applications side and
chokes that off, that will drop how much he can charge on the other side. Now,
nobody knows what the right pricing structure is. I don't claim to know it. Nobody
does. There is no presumption that the right structure is to recover all the costs of
consumer broadband networks from consumers alone. No presumption of that.
What would happen if you did allow them to charge something to search engines,
let's say, that derive their income from advertising. What economics predicts and
it's independent of a monopoly -- it's independent of the degree of competition in
broadband access. The prediction is if you allow them to charge content providers
in their own interest they will not reduce the price to consumers and therefore
encourage penetration. So I'll be happy to flush that out later in the Q & A.
Bottom line, you know, I'm not a reflective anti-regulation person, but I am, I think
it's safe to say, I'm worried, and let's regulate without really thinking hard about
what are you going to do that won't be a disaster is really not a solution. I'm willing
to listen to suggestions but I think we need to, you know, be a little bit more
concrete about exactly what we plan to do.
>>WILLIAM BLUMENTHAL
Marius, thank you. Barbara, all yours.
>>BARBARA TULIPANE
Okay, the last panelist on the last day, or the last hour, so everybody shake it up.
Move around in your chairs. I can see eyes starting to close. I will be quick and to
the point. I want to thank the Federal Trade Commission and Maureen especially.
Although I came kicking and screaming today, to today's panel. I find these things
very frustrating, because I see a lot of talking like this and I don't think it's
productive. I mean, for every statistic that somebody can throw out I can show
you another one that disputes it. So I just don't think it's productive to go there. I
think we need to kind of shift the focus and talk about what we can agree on and
move on from there. I would say, however, that my members are E-retailers.
They sell directly to the consumers and they do that over the TV, internet or the
radio. It's a little bit unique in that they grew up in the cable model. So when we
talk about theory and hypothetical models as we have today, and nobody really
knows what it will look like, you know, I do. I've been there. My members right
now work in a closed network, and it ain't pretty. So I want to share with you some
of their frustrations and what they go through and maybe we can make some -- we
can draw some conclusions from that and so we don't go down that road with the
internet. Before I do that, however, I think it's productive if we separate
broadband, the other application services from the internet. I only want to talk
about the internet. I get it. The telcos need to recover on their investments, they
can do that. Just don't do it on the back of the internet. The E-retailer
membership started as a community for those selling product on the cable
networks. Cable is an example of closed network. Carriage is not guaranteed.
Pricing is both arbitrary and discriminatory and businesses must negotiate with
regional cable providers throughout the country to reach their consumer. Now, I'm
going to walk you through the steps of what somebody has to go through to get
carriage. Take a company like QVC or HSN. What they must do to reach their
customer. Step one, they have to get carriage. The negotiations begin with each
regional cable network provider. Each has the power to decide if they will add the
live shopping channel to their programming mix. If they are successful with step
one then they move to step two, which is pricing. Pricing can be based on the
number of cable subscribers and/or percentage of sales. In most cases, my
members have to pay both. And it doesn't stop there. The live shopping company
must then negotiate with the cable provide over channel placement. The winners
get placed next to premium content while the losers are regulated to outer space
or the higher channels. Often these decisions are impacted by existing deals,
other live shopping companies have already negotiated. That's how a closed
network works for retailers. Carriage is not guaranteed and companies can be
discriminated against by being placed in the slow lane or higher channels. These
decisions determine if a live shopping company can survive. Within the last two
years, we have seen two large live shopping channels with revenues of over $100
million cease operations due to problems associated with the closed network.
This problem is not exclusive to live shopping, however. As other small content
providers struggle with the same deal making. The model is this. Large players
like ESPN are paid for their content, leaving smaller players to foot the bill. And
that's if they can get carriage at all. We've all seen the commercials. If you would
like a certain program, call your local cable provider. Consumers are essentially
told, here's the deal. Here are your 500 channels. We've packaged it with what
works for us and what makes us the most money. Hardly a consumer friendly
environment. I share this with you because the cable model, and this is an
example of how a closed network prioritizes content. It is solely based on the
network provider's ability to maximize its profits. Contrast this closed cable system
to what an E-retailer experiences today on the open internet. Today, a retailer
simply has one business relationship to gain access to the world as does the
consumer. What more needs to be said. Under open network scenario, the
internet has thrived. Today, however, it is under attack by those that are building
broadband networks. They claim they need to recoup their cost though I would
argue they have already been paid through public business and incentive
programs. We do not disagree that they should be able to sell additional
broadband services like video and phone. It is important, however, to distinguish
between these applications and the internet. The internet is comprised of
interconnected networks that do not distinguish service based on source or
content. As such, I want to be very clear, as we discuss net neutrality today, I am
not advocating for regulation that represents the return to the old rules for
providers broadband offerings as some might suggest. Rather, I am making the
case that basic rules needed for the internet much like those recently agreed to by
AT&T, as it sealed the deal with Bell South. We applaud AT&T for taking this step
and helping to define net neutrality. This position keeps the internet as open
network. Where my small retailers can continue to provide consumers with the
content the consumer wants. Because in today's world, the consumer is in
charge. In fact, there is currently a revolt against top down forced content so why
would network providers want to model their next generation internet in a similar
fashion as the closed cable model? We often hear that someone has to pay for
the additional capacity providers are offering on the internet. Those in my industry
have never asked for a free ride. And, in fact, pay their own way. That's right.
They are already paying millions of dollars to access the internet. However, they
are now being asked to subsidize the role -- phone, internet and television. All we
ask is the internet portion of these offerings remain a viable marketplace where
providers can recoup -- with fair rules of the road. We believe providers did
recoup their investment and create additional revenue streams by charging for
noninternet-related broadband applications. In fact, we encourage their efforts to
provide video or television on the broadband, but let's not mix internet with the
broadband applications. In our industry, innovation is the norm. To compete
against large brick and mortar retailers, E retailers have perfected their sales effort
to meet the consumers' changing shopping habits. In other words, they meet the
consumer on the consumers' term. First on television, and now on the internet.
But what's interesting about ER membership, is that the small players today may
very well be a Google or an eBay tomorrow. Their model for success is their
ability to innovate. In fact, innovation is the backbone of our industry. We
encourage the network providers to follow our members' examples. Innovate
rather than dictate. It's no longer enough to build a walled garden and expect
monopoly rents. Today content providers on the internet are second to none
because they have been forced to innovate. We encourage network providers to
take the same path as this is a proven ingredient for success. I hope that we can
separate broadband services from the internet today. Let's talk about the internet.
As we do this, remember, that prioritization based on source or content will result
in a closed network. Just like the cable system today. I would like to thank again
the Federal Trade Commission and Maureen Ohlhausen. (Applause)
>>FEMALE SPEAKER
I have a fan.
>>WILLIAM BLUMENTHAL
Well thank you, to all of the panelists. I'm going to get out of the way for the Q &
A. (laughter) I'm going to stand. Actually, I just want to stretch. I suspect various
panelists would like to rejoin with comments on what some of the other folks had
to say. I know Alan had some thoughts about Walter's.
>>ALAN DAVIDSON
I appreciate it. I'll just -- I would like to hear questions from the audience. I'll just
say first of all, to the extent, the question of -- this misquote from a Google
engineer in Europe, I would just start by saying, that was really not what he said.
There is a letter, actually one of our folks has it here if anybody would like to see it,
that we sent to the Hill in response to the letter that our friends sent to the Hill, just,
I guess, maybe yesterday. You know, it be better instead of digging out random
quotes from Google employees overseas we actually had a conversation about
what we really mean is that we really do believe net neutrality is an important issue
and we value the role that broadband providers play which is what our engineer
said and look forward to working with everybody to have more broadband. I would
just say, I hope that we can stick to a discussion about the problem that we're here
to talk about, which is the broadband market and not all the other markets. Like I
mentioned, I think, they are radically different markets and we could talk about why
but I think everybody understands that.
>>WILLIAM BLUMENTHAL
Did any of the speakers want to comment on Barbara Tulipane's comments, in
particular the proposal to distinguish between broadband and internet services,
and in particular, to recover future investments from the broadband side alone?
>>MALE SPEAKER
Let me make a couple of observations. First is, I'm not sure that there is an
accepted definition or understanding of what the internet is. I know that there is an
accepted definition of what broadband is, which is absurd. Frankly. The notion
that the internet itself, I mean, I think there is a perception that the internet and
partially we're to blame for it because our engineers draw a cloud to represent the
internet. It's not a cloud. It a series of tubes. (Laughter) But there are a series of,
I would say, blunt, not very sophisticated commercial arrangements between
network operators that form the internet. Those agreements, which sometimes
are referred to as peering agreements, sometimes they are traffic exchange
agreements. They have a variety of terms. They are not sophisticated at all, and
they frankly need to be. And I think it is incumbent upon the folks who operate the
internet to show that those commercial arrangements can result in solutions to
these problems. But I think there are opportunities to solve this problem in
particular. Prioritization of traffic through those peering agreements. Because
there is a sense between the eight or so, certainly between the eight or so Tier 1
networks, there is a sense truly of mutually assured destruction and that
encourages good behavior. So as those agreements mature, I think that there is a
possibility that these problems can be addressed.
>>MALE SPEAKER
I would like to respond to Barbara's comment, too. I understand the concern.
Harkening back to a day when the cable industry had a monopoly and video
delivered by wire.
A day when the telephone industry had a monopoly on voice telephone service
delivered by any means. I understand the concern. Today, it's an entirely new
world. I mean, for example, over the air broadcasting, you now get on your cell
phone. And those who are broadcasters are concerned about, is there going to be
a new bottleneck? The cable industry is delivering voice telephone service and
the telephone industry is beginning to deliver video. We have come to the place
where all of these services are being delivered or will be delivered pursuant to a
technology that is basically internet protocol technology. So when John says
what's the internet? Everything is moving towards sort of an IP-based based
delivery mechanism, whether it's being delivered wirelessly, whether it's being
delivered by satellite, delivered by cable wire, whether it's being delivered by fiber
or by a twisted copper pair by the telephone industry. At the end of the day, I think
the basic concern is the traditional (inaudible) concern. Is the market power. Is
there a market where the consumer doesn't have choice? Is the market
contestable. What we see in the internet market is technology has brought us to
the place where the market is competitive. For those of you sitting right here in
this room, you can access the internet by WiFi, because I can tell that it's on in the
room which then connects to a landline or you can access it pursuant EVDO (ph)
to what's being offered by several different providers right here in this market.
Right in this room today, you can access the internet in a variety of ways. You
have a variety of choices. The traditional analysis of whether there is market
power, whether the market is contestable, whether or not there is the power to
control price -- those are the right analyses, and we support an analysis in that
regard. If there is a problem, address it in the traditional way it's done through
trade regulation by the FTC. But let's not engage in hypotheticals or reach out to
problems that really don't exist.
>>MALE SPEAKER
Walter, this is a theme that Barbara raised, and Alan also the gentlemen from
Skype, which is that if we start charging content providers, they need permission
to get on. Then the transaction costs, they will be so big, that just the transactions
alone could stall the innovation process. That's an argument I've heard also in the
case of wireless spectrum. An argument for, let's have unlicensed spectrum, just
to avoid the hassle cost, the transaction cost. What's your reaction to that
argument?
>>FEMALE SPEAKER
Who are you asking?
>>MALE SPEAKER
Walter.
>>WALTER MCCORMICK
I mean, this is the difficulty of dealing with hypotheticals. But let's take somebody
who wants to offer a new business. Lets say that somebody wants to go into the
business of offering some advanced home health monitoring application that's
going to require some level of prioritization in the last mile, that the individual is
offering it as an entrepreneurial endeavor. That the individual wants to charge for
it. But that the individual has to be guaranteed a quality of service in the last mile
offer. And that the individual would like to be able to offer it at a fairly low cost to
the consumer but the only way to do that is to spread the cost to the network as
opposed to doing the device as a last mile device. I think that it would make sense
for an network provider to be able to offer to such entrepreneurs the ability to go
into that kind of service. When you look at VOIP services, VOIP does require
prioritization. I mean, here's a situation where there is an independent
examination by the Washington Examiner of a service that was offered by Vonage
and he didn't like it but now Vonage is offering a device for $199 that you can add
on to your Vonage service that makes sure that when you're using VOIP services,
it prioritizes the packet over others coming from your computer network. So, I
mean, this is the last mile add on, $199. Why do we necessarily have to charge
the consumer $199 if we could offer it to Vonage the ability to build that into the
network as a network service? So again, I think that this kind of flexibility is
important and that there is nothing bad about it when we have a competitive
environment. When consumers have choices, this traditionally the case, when you
go to McDonald's, you can't order a Pepsi. And when you go into Kentucky Fried
Chicken, you can't order a Coke because there are relationships. You go on
Google and you put buy books, you'll have a prioritization that will give you
Amazon because they've done a deal with Amazon. In fact, if any of us want to
kind of envision what prioritization on the internet might look like, I think the
clearest understanding -- of what no prioritization would be is looking at a Google
search page.
>>BARBARA TULIPANE
This is what I mean about not productive conversation because it's such a silly
argument to say that you relate that the Pepsi, Coke -- I mean the reality is that
you can go to McDonald's and if you don't like that they serve Pepsi, you can go
right next door and make a different choice. So that's choice. But let's be very
clear, I mean GAO had a study that came out and said we absolutely have a
duopoly. So there is not a choice. I don't want to get into tit for tat because I don't
think it's productive. What I always see are the arguments thrown at the Google. I
get it. It's hard to feel sorry for a big company. So let's put those companies aside
and let's talk about the average retailer out there and what that will mean to them.
>>MALE SPEAKER
I would just say there is so much to talk about in Walter's example. Starting with
the fact I think that many providers of voice over IP do not believe they need
prioritization in order to offer their service. Including, you know, Google, has a
(inaudible) a voice product that we offer and I can tell you our engineers think it
works plenty fine without prioritization. But really, the example is actually really an
interesting one because think of the mind-boggling complexity now for a small
business that wants to get online but now feels that it needs to enter into some
sort of a carriage agreement with all of these providers out there. The transaction
costs are enormous, because it's not just here in the United States, the internet
has blossomed. You get online, Barbara's retailers get online and their services
can be available all over the world so a small business in a rural part of America
can be offering services anywhere in the world. Now, what do you do? How do
you do this to try to start entering into these agreements not just with the eight or
10 large broadband providers and shrinking here in the United States, but all the
smaller and medium sized ones and all the people around the world. How do I go
and negotiate in Canada, and the UK, in Thailand, in Japan, so that my services
can be seen? I mean, the beauty of the model that we have right now is that there
is one interface for content providers. Applications, edge services, get online and
are available everywhere. That transaction cost, I think, is a really big part of why
there's --.
>>MALE SPEAKER
I just don't understand that because there is not a -- a consumer can access any
website they want. They are not being blocked, impaired or degraded in any way
by any service provider in this country. So --
>>MALE SPEAKER
And we want to keep it that way.
>>MALE SPEAKER
To lay out an entirely hypothetical concern, when if any service provider attempted
to do that the consumer has the ability to immediately shift service providers.
>>MALE SPEAKER
I think there's some disagreement about that. But even beyond that, this is where
we're in agreement that there shouldn't be this kind of blocking. And we welcome
the fact that Walter said this in Congressional testimony. I think it's great that the
companies have said that. What we're hearing and this gets to the nub of the
argument is that prioritization itself, of certain kinds, can be tantamount to
blocking, what happens if you don't pay for the prioritization? Are you regulated to
a degraded service or slower service that doesn't get consumers what they need?
That's the issue we need to keep discussing.
>>MALE SPEAKER
But today there is not a single instance of any prioritization occurring that
somebody is suggesting is bad. So -- so what you're saying is, let's try and now
define, what services may be created in the future that can be prioritized and what
cannot be prioritized. How does government do that?
>>MALE SPEAKER
Let me -- let me make one observation, and then move on to some slightly
different questions, if I may. The one observation, just harking back to a
discussion about five minutes ago. I suspect that most of you in the room are
drawn predominantly from the internet and telecoms community, and not from
either the antitrust or the soft drink community. But having represented Coca-Cola
for many years, before joining the FTC, I would just point out to you that Coke and
Pepsi actually litigated this issue. It was very lucrative for the law firms, and it
went on for about a decade before ultimately getting resolved, I think in some
settlements. But the issue of selectivity and foreclosure is not is not ordained one
way or the other and that's true even in a dualopoly or triopoly situation. I want to
come back to the dualopoly, triopoly point in a second -- one point of
characterization, I have the same question somebody from the floor had passed
up. Is there agreement when we focus on prioritization that it's a last mile issue?
Do all of you agree with that?
>>MALE SPEAKER
I think that we're here to talk about the last mile prioritization issue. There are
other forms of prioritization that broadband providers might engage in that we don't
see as problematic. So, for example, offering local caching, for example, in the
way that Okamy (ph) does. Okamy(ph) is not here and we'll hear what they have
to say about that but I think many of us have said that that form of prioritization, for
example, doesn't create these concerns because it doesn't -- it's not something
that, where providing it necessarily inherently degrades other content in the last
mile at the router level and it also not something, again, we also believe there is a
market for different providers to provide that service.
>>MALE SPEAKER
When you say prioritization last mile, are you talking about all last mile services?
So cable modem, DSL, wireless, satellite, WiFi, WiMAX, and broadband over
power line? Any prioritization in any last mile service would be subject to
regulation?
>>MALE SPEAKER
I think we've already said, and many other folks have said, that there are lots of
different kinds of network management that is not what's at issue here, right? And
also I think it's fair to say what people are worried about is anti-competitive
prioritization. That's really part of the thing, the discussion. I would note, you
know, that there is quite a bit of disagreement about the level of competition.
You've painted a picture that I think most people would argue is not entirely
accurate. (overlapping speakers)
>>MALE SPEAKER
I'm just asking, with regard to -- would you treat all last mile exactly the same and
make it subject to some sort of government regulated approach to what traffic
would be prioritized? For example, let me just, on your wireless phone, you
receive, on my wireless phone I receive an email, I receive -- I can access the
internet, I get voicemail. Would there be some regulation with regard to what
receives a priority in email? (overlapping speakers)
>>ALAN DAVIDSON
As you know, there is a great deal of, you know, question about these different
markets and about how much competition there is there and whether there are
differences. We've been talking about the last mile and the wire line context,
where I think many of us agree, the biggest concerns are. There are others who
talk about the wireless environment. And the extent to which there are issues
there. And --
>>MALE SPEAKER
Specifically, what are you advocating, Alan? (overlapping speakers) Specifically
what it would be?
>>ALAN DAVIDSON
Right.
>>MALE SPEAKER
The government would say, with regard to last mile services, which services would
be covered?
>>ALAN DAVIDSON
I think we said, we've said that it the last mile services.
>>MALE SPEAKER
Of all last mile providers?
>> ALAN DAVIDSON
Yeah. It's very much similar to what was in the AT&T agreement, merger
agreement, and that's exactly the sort of approach that all of us have said. It's
quite simple. It's really, it's one sentence. It the notion that, there cannot be this
kind of discrimination in the last mile based on the source or content of
communication. It's been, it's been very simply put by the FCC in that merger
agreement. We would be the first to say, there might be multiple ways that you
can get at this problem and that's why it's great to be here talking to the FCC
about this but it actually an extremely set of things and the only reason I use these
examples is simply to give everybody some sense, we're not talking about some
massive regulation of the internet or some kind of regime that people weren't living
under until about a year and a half ago anyway. I think we're talking about
something that's very simple, it's not a heavyweight kind of regulation. And it's
aimed at a very particular set of practices.
>>MALE SPEAKER
One more question and I'll shut up. I still don't understand the something but if I
have some concept of it, then a last mile provider by WiMAX such as Google is,
and Google will offer WiFi, WiMAX access for free if you agree to take a prioritized
delivery of advertising from Google --
>> ALAN DAVIDSON
Our network is offered in the neutral way. Our WiFi network in Mountain View.
We would encourage others to do exactly the same thing.
>>MALE SPEAKER
When I call it up and I get an ad that's a prioritized delivery that that advertiser is
paying to Google?
>> ALAN DAVIDSON
I don't see where that's a prioritized delivery. Honestly. We offer --
>>MALE SPEAKER
The first one I receive over the WiMAX network
>> ALAN DAVIDSON
I think -- I would like to know what you're talking about because honestly.
>>BARBAR TULIPANE
Can we move the Google and the USDA conversation to another time? It always
ends up to be about Google and I think that's a switch and bait and not productive.
>>WILLIAM BLUMENTHAL
Fair enough. I was actually going to say Walter, the same question back to you,
with a slight variation. Just to clarify, is it -- I understand that your position is that
there are enough sources of competition right now that the issue ought to be moot.
But if there were not enough sources of competition, if there were, say, a single
provider, in a particular locality, would you, in that case, acknowledge the
legitimacy of the concerns that the neutrality --
>>WALTER MCCORMICK
I would say several things. First, is the market contestable? Second, is there a
real, definable problem that merits government intervention? And third, what are
the ancillary cost of dealing with that particular solution that's being offered? In
this instance, I can't quite get a handle on exactly what the problem is. Number
two, there is competition, and number three, the market is clearly contestable. I
mean, so that, I think those are the --
>>MALE SPEAKER
I'm not sure if that was a chuckle of support or not. One way or the other. Let me
just ask two follow ups to just make sure I understand. First, for purposes of
market share calculation, since you're identifying all of the different technologies
that presumably are hitting this building one way or the other, the bulk of the traffic
out of the building is going on one particular one, for purposes of market share
measurement is it your view we should be doing a one over end analysis where all
the different technologies are given equal share or for purposes of measuring
share should we actually look at the amount of traffic going over the different
modalities?
>>MALE SPEAKER
I think if you're going to start defining the relevant market, you know, you have to
begin with what exactly is the market that we're looking for. I think that if --
>>MALE SPEAKER
Assume it's broadband services?
>>MALE SPEAKER
It is last mile broadband access, the FCC's own statistics show, that the growth in
that area which was 26% growth in last mile broadband connections in the first six
months of '06, shows that 58% of those new connections were wireless.
(overlapping speakers) What I'm saying is you do have a variety of providers.
>>MALE SPEAKER
I think you've got to look at what most people have access to. I mean if you look
at the broadband statistics and we'll hear all about them tomorrow. They are
terribly flawed but 99.6% of Americans are getting their broadband access
through their incumbent cable or wireline telephone provider. And 34%, almost
34% of Americans only have one option for broadband provider. 13% only have --
have none. That's from the latest statistics, okay? So there will be a whole panel
tomorrow to discuss this part but I think you've got to look at what kinds of stuff
people actually have access to.
>>WILLIAM BLUMENTHAL
We'll sort out the facts. I was just trying to get -- to the final issue, is it one over
end or actual analysis. For those from the antitrust community, know that's a
pretty familiar type of analysis so we'll look at that. Walter, one other question I
had for you on market share measurements, to the extent that the providers of the
wireless broadband services are subsidiaries of the same companies, that are
providing the DSL, typically, we would aggravate those, do you have any issue
with that, or do you think those should be disaggregate?
>>WALTER MCCORMICK
I think that the way in which a lot of those services are now being marketed, is as
alternatives, so I think it should be disaggregated at this point in time, because
what we're talking about are completely alternative services for purposes of last
mile internet access. And I think that what we're going to see in the future is we're
going to see even greater kind of mixes of services being offered both wireless
and -- wire line and wirelessly. I think this disaggregated for now.
>>WILLIAM BLUMENTHAL
Let me ask one last question, it may be the last question we have time for this
afternoon. I wanted to come back to the two-sided market issue. And in
particular, it was triggered in a bunch of places. Barbara's comments about the
difficulty of getting carriage triggered it but much earlier, in my notes from the
various speakers, back when Alan was speaking, I jotted a note down to myself,
that simply said settlement mechanism. Question. And the basic intuition is, when
you look at two-sided markets, you know, blanket licensing by (inaudible) and BMI,
stock exchanges, MasterCard and Visa, there are well-established mechanisms
for figuring out how financial settlements are going to occur. So that you don't
have to have 80 different contracts or 80,000 contracts. You simply have one
payment from any given player. To the panel as a whole, to the extent that people
are thinking about, about some surcharges, or some selective charges, other than
to one of the players at either point. Somebody who is not in privity with one of the
players at the end would have a mechanism for surcharging somebody who is not
their historical customer. What's the settlement mechanism y'all are thinking
about?
>>BARBARA TULIPANE
We're not. (laughter) (overlapping speakers)
>>MALE SPEAKER
I don't think anybody is actually. That was one of my observations, about how
intercarrier compensation works or doesn't work on the public switched telephone
network. It's horrible. And it's highly regulated, but still, horrible. I don't think
anybody -- even the incumbents who have indicated that they have a desire to
eventually charge for priority access, we sit down and talk to our engineers and
say how would it work. And the answer is, not very well at all. It's not particularly
feasible to implement and I'll date myself and give you a sense of what my TV
watching habits were as a kid, I used to watch "Battle of the Network Stars." I
think you'll have a gargantuan battle of the network engineers the moment that
any one of the incumbents attempts to charge for priority access. And it's not that
hard get around any of the priority schemes that you can envision being
implemented at least at this point in time.
>>MALE SPEAKER
I agree. I think it's extremely difficult to imagine how would you do this especially
internationally. And, you know, I also wonder if we really know which way all of
those revenues would flow. Whether it wouldn't be content providers ultimately
end up charging to make sure their content is seen. I just think we don't know
about it so we're left with this kind of miracle of the network that we have right now,
which is where people are able to get online by paying and paying, you know,
quite a bit to their own service provider, to get access to the network.
>>WILLIAM BLUMENTHAL
We're running long as it is. So with that let me draw this panel to a close. To
those of you who sent forward questions that we didn't have a chance to get to, I
apologize for that. I'll make sure we have them in the hands of the organizers for
purposes of tomorrow's panels, where I suspect these same the issues will come
up perhaps with a slight twist. Let me ask you to join me in thanking the panel for
their time. (Applause) Maureen, I'll turn it over to you for some housekeeping
details.
>>MAUREEN OHLHAUSEN
For those who are wondering about tomorrow, we will follow the government-
closing schedule. So if the U.S. Federal government is closed, the workshop will
be postponed. If the government opens late, we will open late, accordingly. So if
it's one hour late or two hours late we will follow that schedule. You can find that
information on the site OPM.gov. And we'll have the Federal government's
Washington, D.C. operating schedule. Thank you so much for coming and I really
hope to see you all tomorrow. (Applause)
- See FTC Workshop Transcript for index
