We have carefully reviewed the proposed modifications of the International Telecommunication Regulations (ITRs) contained in TD-64, as well as some other recent proposals and some ITU presentations made in the runup to WCIT. This review confirms that the most important potential effects of the ITRs on the Internet would come from the ITRs’ attempt to change international Internet connectivity arrangements in fixed networks. For a very useful summary of ITU work on tariff and accounting matters, international mobile roaming, international Internet connectivity, and taxation issues, see this presentation from a February 2012 WCIT preparatory meeting in Bangkok. The motivations of these proposals are primarily economic; they have to do with the flow of funds and the role of national regulators in controlling operators and commercial negotiations, rather than censorship or “taking over” Internet resources.
We have gotten into some disputes with well-intentioned advocacy groups by insisting on the disconnect between censorship and the ITRs. (See the comments on this post) While it might be politically convenient in the U.S. to paint the ITR modifications as a plot to control or limit the free flow of information, we academics have to insist on precision and on perspective and context. We understand that some kinds of threats “mobilize the base” better than others, but we don’t think anyone benefits in the long run by misconstruing the nature of the problem. Indeed, such distortions might backfire. Changes in charging arrangements and economic regulation could have very important effects on the free flow of information. If we are to understand those threats and how to handle them we need to accurately identify what is being proposed, why it is being proposed and what its effects might be.
Summary of what’s there
The word “Internet” appears in six proposed modifications of the ITRs in TD-64. There are a few other proposals that do not refer to the Internet directly but which could affect it. Certain definitions, such as those pertaining to spam or security, could also have an impact, but the direct references are the most important ones, and most of them concern the economics of international internet connectivity.
The first mention of the Internet comes in a proposed change to section 2.2, which defines international telecommunication service. It would extend the definition to include “Internet traffic termination.” (There is a counter-proposal to simply delete section 2.2 entirely.) A similar proposal to modify section 4.2 would add “services for carrying Internet traffic and data transmission” to a laundry list of services that members states agree to cooperate to help provide.
A second direct reference to the Internet comes in a proposed new section 3.7. The ITRs would ask administrations to:
take appropriate measures nationally to ensure that all parties (including operating agencies authorized by Member States) involved in the provision of international Internet connections negotiate and agree to bilateral commercial arrangements, or an alternative type of arrangement between administrations, enabling direct international Internet connections that take into account the possible need for compensation between them for the value of elements such as traffic flow, number of routes, geographical coverage and cost of international transmission, and the possible application of network externalities, amongst others.
This is similar in intent to the proposal by the European Telecommunication Network Operators Association (ETNO), which was released publicly by the ITU at the request of ETNO. The ETNO proposal:
- Brings “IP interconnection” fully into the domain of the ITU by defining it in the ITRs
- Creates another two new definitions for “best effort” and “end to end quality” for the delivery of “packet data units” – in other words, packet switching becomes officially part of the ITRs.
Based on these definitions, the ETNO proposal then tries to do two things. First, it tries to ensure that member states allow interconnection arrangements designed to allow end to end quality of service as well as best-effort packet forwarding; second, it urges operating agencies involved in negotiations “to ensure an adequate return on investment in high bandwidth infrastructures,” by making their commercial agreements “achieve a sustainable system of fair compensation for telecommunications services and, where appropriate, respecting the principle of sending party network pays.”
The next direct reference to the Internet comes in a proposed new section 6.7. One of the most interesting proposals, it says:
6.7 Member States shall ensure that each party in a negotiation or agreement related to or arising out of international connectivity matters including those for the Internet will have standing to have recourse to the competition authorities of the other party’s country.
As far as we can tell, this means that international peering or interconnection negotiations would allow an unsatisfied party to drag the competition authorities of the other party’s government into the negotiations. One can see why certain large U.S. companies might not like this, but one can also see how it might allow competing companies, from the U.S. or elsewhere, to challenge state-owned or private monopoly international gateways with appeals to local competition authorities. Another proposal modifies the above by adding “will have access to alternative dispute resolution mechanisms.”
The last proposed ITR change that directly references the Internet is a proposed new section 8.A.4 which would say,
8A.4 Member States shall take measures to ensure Internet stability and security, to fight cybercrime and to counter spam, while protecting and respecting the provisions for privacy and freedom of expression as contained in the relevant parts of the Universal Declaration of Human Rights.
Nice. The words here are unobjectionable in their literal meaning, but typical of the vague, impossible-to-operationalize language that often populates international agreements. We will have more to say about the intersection of cybersecurity issues and the ITRs in the next blog post in this series.
Another proposed revision that clearly targets the Internet, without mentioning it, would revise 3.1 of the ITRs to require operators to
cooperate in the establishment, operation and maintenance of the international network to provide a satisfactory quality of service [and above a minimum level corresponding to the relevant ITU-T Recommendation].
The bracketed language (which indicates a lack of support) is a straight-up protectionist measure. A mandatory minimum quality of service would have the effect of banning many Internet-based services and applications, because so much of the Internet relies on best-effort packet forwarding. For example, a Voice over IP telephone service might often fall below regulatory minimums regarding quality of service. Consumers often prefer low quality anyway, because they are willing to trade quality for lower cost (e.g., Skype).
It is clear that several proposals revise definitions to make Internet, and especially “international Internet connectivity” a subject of the ITRs. Obviously, this continues the old debate about what is telecommunications and whether it includes Internet, as described in our last blog. Defining Internet as a telecommunication service is not something we favor, because it would pave the way for regulating what are now privately negotiated, contract-based interconnection arrangements subject to market forces.
All the proposals discussed above show how many of the ITR modifications are not new attempts to “take over the Internet” but a continuation of long-running battles over the way the Internet has disrupted traditional telecom businesses and markets. The proposals reflect the desire of certain companies that operate layer 1 and 2 networks to be better compensated for carrying the burgeoning amount of data traffic at the application layer, or to be protected from it. For some administrations, mainly developing countries that might still have monopolies or state-owned champions running their networks, it reflects dissatisfaction with the Internet model of paying for your own connectivity to the Internet (sender pays), rather than the old shared-payment model, and some dissatisfaction with contractually negotiated as opposed to government-regulated interconnection.
The battle over international internet connectivity is not new; it goes all the way back to 1999, when various countries remote from the centers of Internet content, including for a while Australia, complained to the ITU about the new Internet model of paying for connectivity rather than sharing the cost of international bandwidth with your corresponding country operator. This resulted in ITU-T recommendation D.50, first passed in 2000 and since revised several times. D.50 is just a recommendation; some of the proposals try to make it more of a requirement by bringing it into the ITRs. This may be less of a threat than it seems, although it bears watching. The ITU has done endless studies and engaged in endless wrangling about how to measure and charge for Internet traffic. It has never come up with a solid, consensual proposal on how to do this. Just as AT&T’s early effort to get YouTube and other application-layer players to pay more for the tubes never went anywhere, it is not entirely clear how ETNO’s idea of “commercial agreements to achieve a sustainable system of fair compensation for telecommunications services” would be implemented. ETNO is also asking for a free pass to negotiate end to end quality of service interconnections; this should give net neutrality advocates pause, but national regulation could ensure that it is not used to discriminatory effect.
This entire effort, in our opinion, should be abandoned and not enshrined in the ITRs. ETNO’s members already can negotiate any charging arrangements their partners will go along with. If they want regulators involved it must be because the market will not deliver what they want. The idea that connectivity charges should be rigidly governed by some regulatory-mandated metric like duration in minutes or traffic or number of packets or the direction of packet flows is simply an anachronism that does not reflect the heterogeneity of online services and the prevalence of multi-sided markets. Data is not voice, and telecommunication authorities pining for the good old days of predictable settlements simply need to wake up and face the facts about the way Internet applications work.
We have so far avoided the topic of international mobile roaming charges because it has traditionally been an area that could be unambiguously classified as telecommunications, at least when voice communications are involved. But we all know that data communications are taking over the mobile sphere just as they did in fixed telecommunications. And we also know that international data roaming charges are, if anything, even more outrageously overpriced than voice roaming charges.
As far as we can tell, the proposed ITR interventions focus on standard consumer protection matters such as transparency of prices, notification, and affording consumers an opportunity to decline additional paid roaming services. There does not seem to be any effort to regulate or control the charging system for mobile data interconnections specifically; indeed since most of the mobile data that crosses international boundaries probably does so through a fixed network, it is the fixed network charging arrangements that one has to watch. There is a new section 4.6 that tries to regulate the quality of service offered by mobile roaming agreements; this has little support however. The most salient section is this:
4.4 Member States shall ensure that operators providing international telecommunication services, in particular international roaming, provide transparent and up-to-date information on retail charges, including roaming charges. [In particular, each customer should also be able to easily have access to, and receive appropriate and timely pricing (including taxes) information free of charge when abroad on the relevant price plan, except when the customer has notified his home operator that he does not require this service].
There are several efforts in the ITRs to extend standard forms of telecommunications regulatory authority over international internet connectivity arrangements. These economic interventions could be damaging to the Internet’s status as a relatively open platform for new services if passed, but it is unclear how much support they have. The most potentially disturbing proposal is that of ETNO, because it shows that major developed-world telecom companies want to incorporate international internet connectivity into the ITRs. But even their proposal would be limited in its impact because its proposed new charging arrangements, calling for “sustainable” and “fair” compensation, are so vague. It would, however, be a very bad long-term precedent to incorporate such things under the ITR if one believes in a market-driven, free and open Internet.
Next: WCIT and cybersecurity