Listen to this article
Ricardo Perissich is the head of regulatory affairs at Telecom Italia and chief executive of Telecom Italia Media. Mr Perissich says network neutrality is an issue for European telecoms companies as well, and he has raised this issue with the European Commission.
Andrew McLaughlin is senior policy counsel at Google with a long association with public internet policy. He was involved in setting up the Internet Corporation for Assigned Names and Numbers (Icann), the global internet governance body, and held several senior positions there between 1998 and 2002.
Mr Perissich and Mr McLaughlin answer readers’ questions below on net neutrality.
I don’t think it`s appropriate for the government and authorities to decide the product price for a private company. Some say, however, considering some major telecoms advantages in brand recognition, customer pool and bandwidth use, the government and authorities must have the right to regulate the company`s call rates to ensure fair competition . Could tell you me about it?
Shin Park, South Korea
Andrew McLaughlin: I agree. Net neutrality has nothing to do with price controls. I oppose price controls, and they are not relevant to the debate. Carriers should be able to price according to the market: they should be able to charge end customers flat rates, volume rates, or whatever. Likewise, global interconnection rates should continue to be unregulated. The existing market-driven system works well.
Ricardo Perissich: I agree. After eight years of a liberalised European market, we now have a situation where there is competition and this has been recognised by the Commission. Having said that, we recognize that that there are reasons to retain some sort of regulation on intermediate services, particularly where there are identifiable bottlenecks in the networks. But we also think that all operators should be given freedom, subject to antitrust controls ex-post, in the provision of retail services. This is primarily in the interest of consumers who want innovation and new services at the best possible prices.
From a strictly legal point of view, regulation is indicated where a single operator is seen to have a dominant position but as that dominance is progressively whittled away the extent of regulation diminishes proportionately too.
For sensitive traffic, such as high quality video, how far into the network does Google think net neutrality should extend? (Starting at the customer: Just the local loop, or all the way back to Google’s nearest point-of-presence, or somewhere in between?)
AM: Net neutrality is an admittedly vague shorthand for two important notions:
1. Last-mile Broadband Chokehold Abuse: Where a broadband provider holds a dominant market position as to some group of end-customers, it should not leverage its dominance at the network layer to advantage particular competitors at the service layer. In other words, where there is not a well-functioning competitive market for high-speed broadband connectivity (e.g., a de facto monopoly or duopoly situation), the available dominant carriers should be required to operate their networks in a nondiscriminatory manner.
2. Global Interconnectivity: The Internet should continue to be a network where every end point can reach every other end point. Individual users, not carriers, should determine what sites and services they wish to access. This means that broadband carriers should continue to deploy the global technical standards that define the Internet, and not replace them with proprietary protocols that effectively turn their networks into walled gardens.
So given that, I’d say that network neutrality Type 1 (Last-mile Broadband Chokehold Abuse) is most important at the last mile, while net neutrality Type 2 (Global Interconnectivity) is most important across the Internet.
Mr Pessirich, how do you think that a tariff scheme such as that proposed by the telcos could be transparent and fair when they play the tripleplay game?
Daniele Della Seta, Rome
RP: We think that customers want simple tariff schemes and with convergence operators are moving to flat bundled offers, which will become the rule. So, consumers, rather than choosing between single service offers, will choose from comparable bundled offers.
1 There is increasing competition in network provision. Restricting operators’ charges is regulatory intervention into a fast developing market. Regulatory intervention requires justification on the basis that it is reasonably foreseeable and likely that the market will fail to deliver results as effectively as with the regulatory intervention. What are the panel’s views on how forseeable and likely this is?
2 The importance of proliferating public-interest content (e.g., online news services) may merit regulatory intervention, but a vast amount of content is commercial entertainment and it is hard to see why it merits the free ride. Might not allowing operators greater freedom in charging permit more efficient investment resulting in better networks as well as removing pricing distortions from the provision of content itself?
3 Lastly, where public-interest content merits prioritization, might an alternative arrangement apply, such as the must-carry obligations that are used with cable operators, or special frequency allocations made to public broadcasters?
Rory Macmillan, Geneva, Switzerland
AM: 1. There is no reason to restrict operators’ charges. Net neutrality is not about price regulation. Rather, it is about “reasonable non-discrimination” -- meaning that last-mile broadband operators should not abuse a dominant market position by privileging some among many competitors in the higher content/service layer. The need for regulation comes from market failure, so I would agree that a fully competitive last-mile broadband market -- meaning one where there are many avenues of high-speed broadband into the home, such as fiber, cable, phone, wireless, and powerline -- would not need net neutrality rules.
2. There should be - indeed, there are currently - no free rides. All content providers, whether public-interest or for-profit have to pay for connectivity. Carriers are free to charge their customers what they wish. The question is whether carriers with a chokehold over some group of users (due to lack of meaningful broadband competition) should be able to extort monopoly rents from content providers who are customers of other carriers.
3. The internet is a world where packets are packets are packets. It places the power to distinguish between packets in the hands of the network’s edge -- its individual users -- rather than the intermediary carriers. That edge control is what has made the internet such a powerful and fast-growing platform for innovation. Network operators should not distinguish between different kinds of content. At the scale of the global Internet, it is impractical for carriers to distinguish between “public-interest” or any other sort of content. Nor should they. On the internet, network operators have one job and one job only: move all packets from source to destination as rapidly and efficiently as possible.
RP: We recognise the importance of delivering public interest content through new media and are, of course, more than willing to discuss the subject with the authorities. In some areas of public service, for instance health services, new technologies could lead to significant cost reductions.
Mr Pessirich, what will a feasible net-access charging model look like? Will we see the birth of a sort of telco-guild, collecting fees from the content providers’ one?
RP: We believe that telecommunications operators and content providers have a common interest in offering the maximum amount of content to the market at the best possible prices. So prices have to be attractive and the model, we think, should be based on revenue sharing systems, and easy, fast methods of acquiring rights; in Europe the present system is too fragmented. We also believe that revenue sharing systems should not be regulated but are best left to commercial transactions between telcos, service providers and content providers. It is important to keep in mind the different natures of the digital world and the one that preceded it which you could call the Gutenberg world. In the Gutenberg world (not only bookshops, but, movie theatres, CD and DVD shops, even traditional TV channels) you have physical constraints on distribution and a consequent restriction on available space for content in the distribution system, which, in turn, lead to a privileging of high-margin, premium content. In the digital world you don’t have the physical limitation but the possibility of an almost infinite offering and the so-called long tail phenomenon, which means that a wide offering, including low margin content, is not only economically feasible but an opportunity, as illustrated by cases such as Amazon and ITunes.
One argument against net neutrality is that a legislative approach could be be too prescriptive. How would enshrining net neutrality in law avoid posing problems of its own for internet innovation? Doesn’t the nature of the internet - both from a technical and business point of view - mean that it is best left unregulated?
AM: This is a crucial question. Nobody wants the Internet to be a heavily regulated network. If we define net neutrality as “reasonable nondiscrimination”, rooted in basic principles of competition law (e.g., a dominant market player should not be allowed to abuse its dominance in other markets), there is no need for detailed regulation of the network. Rather, the role for government is to make a clear statement of the “reasonable nondiscrimination” rule for network operators, and then be prepared to enforce it. There is no need to regulate prices or network protocols, for example.
Can you tell us more about how seriously Google is taking the threat of higher charges from telcos? Reports and rumours about Google’s “dark fibre” resources have prolifered in the last year or so - is Google hedging its bets by building out more of its own infrastructure?
AM: Net neutrality is not about carrier charges. We pay large sums of money to our carriers for connectivity to and from Google’s data centers. If our carriers raise their prices, we can switch carriers. That kind of choice does not exist in the vast majority of residential broadband markets. Net neutrality is about whether broadband providers with a chokehold over the last mile of connectivity into the home can discriminate by privileging or degrading particular websites and services.
Next Generation Networks will support multiple Quality of Service levels going from ‘best effort internet’ to premium QoS level suitable for real time video. Surely content providers should pay for higher QoS to give a reasonable rate of return commensurate with the carriers investment to support these service levels?
David R Smith, UK
AM: I have no problem with network operators implementing QoS for, say, real-time video. And they should be able to charge for it. But where there is not a well-functioning market for last-mile broadband -- where a home has only one or two options for broadband connectivity -- the provider should not be able to grant QoS to one competitor while denying it to another. QoS capability should be offered in a reasonably nondiscriminatory way. “Reasonably” is a key word in this context -- a network must always be able to discriminate among packets where there is a good reason, such as stopping a distributed denial of service attack. But it is not reasonable for a broadband provider to allow one competitor to pay for QoS, while another is denied.
RP: We agree with Mr McLaughlin in principle. We think that telcos should be allowed to handle quality-of-service on a commercial basis, but in a fair and not discriminatory way. We would be opposed to legislative or regulatory intervention limiting the freedom of commercial partners. Antitrust law should be enough.
Telecom companies charging for usage seems to be a slap in the face for their potential leadership role in the convergence of media, tech and telecoms.Or is it that they feel they are bound to be losers in the race to add value (and deliver content) and so see their future as a utility merely metering and charging for bandwidth usage?
Philip Letts, www.businessuncut.com
AM: Telecom companies have assumed dominant market positions in the provision of broadband connectivity to many cities around the world. They are welcome to move beyond commodity connectivity to enter higher-layer services like voice, video, web content, etc. What they should not be allowed to do, however, is to unfairly and abusively leverage their dominant chokehold over last-mile connectivity to advantage their own higher-layer services at the expense of competitors. That’s the thinking behind the “reasonable nondiscrimination” principle that we call net neutrality.
RP: In this game, if we succeed in offering consumers what they want, there should be no losers but telcos, service providers and content providers should all gain. That is our approach. Telcos have to invest heavily in the networks. Service providers must produce innovative interfaces and content providers must produce attractive content. All these efforts must be rewarded, otherwise the market will not develop and that will be to everybody’s loss.
AM: At the risk of identifying common ground (a debate host’s nightmare, I know -- I promise to compensate with a rowdy food fight in a moment), it seems that Mr. Perissich and I agree on a key point:
“From a strictly legal point of view, regulation is indicated where a single operator is seen to have a dominant position but as that dominance is progressively whittled away the extent of regulation diminishes proportionately too.”
I would only disagree with the word “single”. As we have seen in the case of mobile phone operators in the US, the existence of multiple competitors does not automatically lead to open, user-empowering networks. In my view, a well-functioning competitive market in last-mile broadband connectivity requires at least three competitors capable of relying supply of high-speed Internet.
We partially agree on this: “...there are reasons to retain some sort of regulation on intermediate services, particularly where there are identifiable bottlenecks in the networks. But we also think that all operators should be given freedom, subject to antitrust controls ex-post, in the provision of retail services.”
I think that a version of ex-post antitrust controls is probably the best general enforcement approach. But regulators should make clear the requirement for “reasonable nondiscrimination” by operators with dominant market positions at the last mile. Such operators can deploy new fast-lane services, and can charge both customers and providers for them, but they should not be able to grant special access to some competitors while denying it to others.
RP: I hope that Mr McLaughlin’s focus on the word “single” does not imply an inclination to use the concept of joint-dominance extensively. That is a highly controversial concept under antitrust law, at least in Europe. The definition of conditions to assess the degree of openness of specific markets is a very complicated process that should be left to regulators.
Content providers already pay extortionate fees for bandwidth. How would these proposals differ from the existing situation and how, practically, would fees be levied?
James Manktelow, London
AM: Presumably, what’s contemplated is the last-mile deployment of (a) proprietary protocols, and/or (b) reserved network elements. Either would create an artificial chokehold that would force online content-and service-providers to have to cut a deal with the carrier to reach its customers. For ordinary Internet traffic, such a network would be a crummy one, turning the Internet from a global edge-driven platform for innovation into a loosely-joined collection of centrally-controlled walled gardens. If the Internet teaches nothing else, it has shown that the old telecom model of central control is vastly inferior to the end-to-end, power-to-the-edge, individual-in-control architecture of the Internet. We should be very cautious about abandoning that lesson.
RP: If you think that content providers already pay extortionate fees for bandwidth, you have never negotiated with an American major!
The success of the content providers is it not dependant on the quality of service provided? If yes, should not there be a charge by the providers? In the end is it not this the way that a free market is supposed to operate?
AM: I have no problem with network operators charging extra for guaranteed-quality-of-service functions. What Google opposes is network operators making those functions available only to some competitors, rather than to all on a reasonably non-discriminatory basis.
RP: Interpreting the future in the way Mr McLaughlin does is highly speculative, particularly in view of the fact that the architecture of new generation networks (NGNs) is is being designed as we speak. Clearly, telcos, in any event our company, are perfectly aware of the nature of the Internet and will not want to crimp its growth in anyway. And as I said in answer to an earlier question, it is in everybody’s interest that the market develops.
Should regulators intervene? Is the solution to promote the use of innovated network access such as unlicensed spectrum? Should regulators encourage big providers to build their own infrastructure on the internet? But then, what to do with the small / innovative internet start-ups?
Maximiliano S. Martinho, Brazil
RP: Regulators should act in a way to encourage investment between competitive platforms. In the past, the focus was on the wholesale obligations of dominant operators, but this can only be useful in a transitional phase. This is clearly the line taken by US regulators. It is increasingly the case also in Europe. Spectrum is a separate issue that would require a forum of its own.
AM: The core problem here is lack of meaningful competition in last-mile broadband connectivity. In the absence of a multiplicity of competitive offerings (fiber, copper, cable, wireless, powerline, etc.), there is a need for regulators to ensure that connectivity chokeholds are not being abused to create advantages for one or more competitors among the many who depend on the network to reach their users.
To address the core problem in cases of market failure, regulators should (1) enforce the principle of reasonable nondiscrimination, and (2) encourage new avenues of last-mile broadband access by, as you suggest, unlicensing as much spectrum as possible.
The debate explained
The rapid rise of high-bandwidth internet content, such as video, has led several telecommunications companies in the US to push for higher fees from the content providers in recent months, which could raise costs to not only established internet companies such as Google and Yahoo!, but smaller start-ups.
Advocates of “net neutrality” argue that such a move would go against the very nature of the internet, which they say has fostered innovation because its architecture is based on open and equal access. Charging content providers more, they argue, could stop tomorrow’s Googles from even starting.
However the telecommunications industry has spent vast amounts in its high-bandwidth infrastructure and argues that it should be entitled to a share of profits gained by using its networks. Furthermore, they say that demand for faster broadband will mean even more money is spent upgrading to “next generation” networks.
The position of net neutrality supporters took a blow when proposed network neutrality price controls in the US suffered a key defeat in late April. Meanwhile some European telecoms companies have indicated they will also seek a new charging regime for internet content providers.