Monday, October 31, 2011

A wider role for National Research and Education Networks (NRENs)?

A new working group paper on broadband and science from the ITU/UNESCO Broadband Commission for Digital Development sets out the importance of national research and education networks (NRENs), in the wider context of broadband's role in supporting science and education communities.

Recommendations specific to NRENs include:
  • Research and Education Networks (RENs), the bodies set up in most countries – including developing countries – to manage and maintain e-Infrastructures, should be given high political visibility towards governments, regulators and academia given their role in the transformation of developing economies into knowledge societies.
  • National authorities and the relevant international organizations should promote affordable and fair access to broadband e-infrastructures via the establishment and consolidation of national, regional and global RENs, fostering cooperative environments that bridge the Digital Divide (non-connected countries and regions) and the Geographical Divide (disadvantaged non-central areas).
  • RENs should spearhead technological and service innovation in partnership with industry.
  • Broadband e-Infrastructures should be leveraged for public service, fostering the engagement of RENs in other public sectors such as e-Health, e-Government, e-Learning, e-Innovation and "e-Capacity Building".
The report positions NRENs as having a much broader focus than their traditional role of supporting the higher education and research community. In the UK, this is borne out in the way the JANET network has extended its reach significantly in recent years, through, for example, the provision services to schools via local authorities and regional broadband consortia, as well as engagement with the Cabinet Office Public Services Network (PSN) programme. See this presentation from a recent JANET Strategic Briefing Day for more on this.

Examples from the USA also illustrate this transition. The United States Unified Community Anchor Network (USUCAN) project aims to "provide community anchor institutions including public safety organizations, public libraries, K-12 schools, community colleges, research parks, and health care organizations with advanced broadband capabilities and services." The initiative utilises the capabilities and reach of the Internet2 national research and education network:
"The U.S. UCAN project was established under the auspices of a federal  stimulus grant to Internet2 from the National Telecommunications and  Information Administration (NTIA) Broadband Technology Opportunities Program (BTOP). Using this stimulus funding, Internet2 is acquiring  more than 10,000 miles of fibre optic cable and will build a new nationwide network infrastructure with an unprecedented 8.8 Terabits of capacity using emerging 100 Gigabit per second technology. This new infrastructure will serve as the underlying infrastructure for U.S. UCAN  to offer its services to community anchor institutions nationwide.  The new network which will be built through strong  public-private partnerships, intends to complement  and link together new regional community anchor  networks created through BTOP funding as well as Internet2’s existing regional network members and  network connectors. The goal is to provide the high  performance national networking capable of fully  supporting all 200,000 community anchor institutions across the U.S. - three times as many institutions as the Internet2 Network serves today."
An ambition very much in keeping with the recommendations set out in the Broadband Commission's working paper, extending the reach and benefits of an existing NREN to a much wider community. Another US project of note is the Gig.U initiative, which comprises "a broad-based group of over 30 leading research universities from across the United States...Gig.U seeks to accelerate the deployment of ultra high-speed networks to leading U.S. universities and their surrounding communities." The project issued a request for information (RFI) in September 2011 which set out four goals as the first step towards Gig.U delivering its mission:
  • Promote the deployment of next generation networks across member communities to stimulate economic development;
  • Identify creative approaches to design, operate and finance self-sustaining next generation networks for member communities while evaluating the trade-offs between these different approaches;
  • Gain an understanding of how differences between member communities influence the level of private sector interest in working with any individual community; and
  • Consider ways in which multiple Project communities can work together beyond the RFI process to improve the private sector business case for next generation networks.
Again, lots in common with the Broadband Commission's recommendations in the above. An FAQ response sets out how the project relates to existing US research and education networks:
"The Project is focused on providing broader community connectivity to the member universities and communities. The existing R&E networks provide significant institutional connectivity to all of the member universities. This effort will neither duplicate nor compete with those networks. Rather, the Project will work with the R&E community and others with network facilities in the university communities, to develop new approaches to extending and upgrading existing network assets with a focus on higher speed retail offerings to places on campus that are not served by the existing R&E networks and to the areas surrounding the campuses. This will enable those who work with ultra-high speed networks on campus to be able to continue their work while home and create laboratories of greater connectivity throughout the university and surrounding community."
A slightly different kind of cross-fertilisation than that being undertaken by USUCAN perhaps, but very interesting nevertheless.

Of related interest are gigabit projects like Gig.U participant Case Western Reserve University's Case Connection Zone initiative, Chattanooga's city-wide municipally-owned fiber-to-the-premises network delivering 1Gbps services as well as Google's fibre initiative in Kansas City (some interesting ideas already here), all of which are test-beds and trail-blazers for exploring the capabilities and possibilities of high bandwidth services. More on these in my next post.

Friday, October 28, 2011

European broadband policy update

Three recent developments to report from Europe: further funding for broadband, two consultations on access to telecoms infrastructure and forthcoming guidance on broadband investment models.

First, funding: €9.2bn will be made available from 2014 to 2020 to "support investment in fast and very fast broadband networks and pan-European digital services", as part of a package of measures to "improve Europe's transport, energy and digital networks", through a new fund called the "Connecting Europe Facility". This announcement implements the commitments made by the European Commission in the Multi-Annual Financial Framework proposals announced on 29 June 2011. More in this press release; an extract:
"The funding, part of the proposed Connecting Europe Facility, would take the form of both equity and debt instruments and grants. It would complement private investment and public money at local, regional and national level and EU structural or cohesion funds. At least €7 billion would be available for investment in high-speed broadband infrastructure. The Commission considers that this money could leverage a total of between €50 and 100 billion of public and private investment – i.e. a substantial proportion of the estimated €270 billion of broadband investment needed to meet Digital Agenda targets on broadband. The remaining CEF funding for digital infrastructure would support public interest digital service infrastructure such as electronic health records, electronic identification and electronic procurement."
Some commentary from Neelie Kroes, Vice-President of the European Commission responsible for the Digital Agenda, further illustrating the additional investment it's envisaged these funds will leverage:
"The Facility would stimulate innovation by using innovation, with funding for broadband infrastructure largely in the form of equity, debt, or guarantees. We would competitively engage new players — like non-telecoms utility companies, or construction firms. Where projects were profitable, public funds could be recuperated and reinvested. And, by giving projects credibility and lowering risk profiles, we could leverage other private and public money. Indeed each euro invested in broadband by such innovative financing could leverage gross private investment of between 6 and 15 euros. In concrete terms, this means that the Facility could leverage a total of between 50 and 100 billion euros. Under the Digital Agenda, our target by 2020 is to get coverage for every European to fast broadband of over 30 megabits per second; and to get 50% of households subscribing to ultra-fast speeds over 100 megabits per second. This proposal constitutes a substantial proportion of the investment needed to hit those targets."
But funding can't be deployed to its best advantage if the regulatory environment isn't right. Which provides a  nice segue into the second part of this update: earlier this month the Commission issued two consultations on access for alternative operators to the fixed telephone and broadband networks of established operators. The first concerns non-discriminatory access for alternative operators to the infrastructure and services of dominant telecom operators. The second concerns the way national regulators calculate prices that operators have to pay for this wholesale access (cost-orientation remedies). Both form part of the Commission's aim to create a single market for telecommunications services in Europe, which requires a consistent approach to regulation and regulated pricing.

The consultation on non-discrimination explores what needs to be in place to ensure that operators with significant market power (SMP) don't abuse their dominant position, to create an open and competitive marketplace as far as possible. It makes for interesting reading, setting out the various ways operators can exhibit "discriminatory behaviour" above and beyond the prices they charge for their products and services. For example:
"…vertically integrated SMP operators may be incentivised to design their wholesale products in a way that favours their own downstream retail arms, i.e. by using particular technical standards or other product characteristics which their own retail arm can use as a wholesale input without further costly adjustments, whereas the downstream competitors may incur additional costs in order to be able to make use of such an input. It may, therefore, be argued that it would be important, under non-discrimination aspects, to ensure that alternative operators are appropriately involved in the design process of future wholesale products and have a sufficient ability to influence the decisions regarding particular product characteristics. In this respect, the question arises whether either the national regulatory authority or an independent industry body should be involved in order to steer and mediate this process.”
Clearly a complex, challenging set of issues, particularly as new NGA services are developed. The pricing consultation is intended to stimulate debate on how wholesale pricing should be set to ensure a level playing field across Europe to encourage investment and competition, particularly in the context of the current transition from copper to fibre. We are a long way from such consistency it seems:
"National Regulatory Authorities (NRAs) are still applying divergent approaches when remedying market failures; in particular, when setting cost oriented wholesale access prices. Even where NRAs apply the same cost model for the same access products, there are divergences in terms of implementation. This leads to a variety of access prices across Europe. The price for the local loop, for example, ranges from 5.21 €/month in Lithuania to 12.41 €/month in Ireland. Consequently there is a lack of predictability and legal certainty for (cross-border) investors, alternative operators and potential market entrants. This constitutes barriers within the internal market. The resulting lack of competition also harms European consumers who are not able to benefit from the choice of services and affordable prices which could otherwise exist.”
Some interesting commentary from Neelie Kroes in her speech accompanying the publication of the two consultations:
"...we need next generation networks to deliver the bandwidth-hungry services and applications that will drive future growth. No-one disputes that need. But there is no agreement on how to foster the deployment of such networks. Unfortunately, we see that, for the time being, telecom companies are hesitant to commit significant funds to fibre roll-out. Most stakeholders consider that this is largely due to the existence of a competing legacy copper network; and that copper pricing plays a key role for fibre investment decisions. However, telecom operators are divided on the question of how copper access prices affect the incentives for fibre investment. Alternative operators consider that copper access prices are too high given that the assets are largely depreciated. And they argue that, as a result, incumbents prefer to make good, easy profits on legacy infrastructure rather than invest significant amounts in new fibre networks. Therefore, they believe that lower copper prices would create the incentive for incumbents to go ahead with fibre investment. On the other hand, incumbents argue that much lower copper access prices would erode broadband retail prices. And, as a result, make it difficult to charge the higher prices for competing fibre products which would be needed to cover the investment costs and risks. In other words, they consider it would be unattractive to invest in a parallel fibre infrastructure directly competing with a cheap copper network, at a time when many consumers do not yet appreciate the major difference, in capacity and service quality, between the two technologies. I think that there is some truth on both sides. And I also have the impression that, as it stands, it would indeed be difficult to build new fibre networks competing with cheap parallel copper networks."
She went on to outline two approaches to address this apparent impasse:
"First, a general approach that would, in principle and after a certain time, gradually lower the access prices for largely depreciated copper networks. Second, the possibility of derogating from or adapting this general approach where the incumbents credibly commit to invest in fibre networks in a relevant time frame, while at the same time promoting the switch-off of the old copper networks once the next generation infrastructure is in place. Indeed, I have seen evidence that the gradual switch-off of copper could reduce the cost to such a degree that new fibre investments break even in under 10 years. And thus align the interests of investors and long-term financing providers. The approach I have just described would lower copper prices in areas where incumbents' legacy copper networks persist for some time without significant fibre investment. But at the same time, it would create an incentive for incumbents to replace old copper networks with new fibre infrastructure. Clients could then be migrated to fibre, and benefit from better services and applications for which higher wholesale and retail prices could justifiably be charged. Such a mechanism should also reassure markets that investment in fibre is safe and profitable."
A "fibre switchover/copper switch-off" is an interesting idea, especially in the light of the new funding to be available from 2014, as mentioned earlier in this post. The wholesale pricing consultation also makes reference to the Commission Recommendation of 20 September 2010 on regulated access to Next Generation Access Networks (NGA), which contains the following acknowledgement of the challenges in driving NGA demand at paragraph 23:
"The deployment of FTTH will normally entail considerable risks, given its high deployment costs per household and the currently still limited number of retail services requiring enhanced characteristics (such as higher throughput) which can only be delivered via fibre. Investments into fibre depend for their amortisation on the take-up of new services provided over NGA networks in the short and medium terms. The costs of capital of the SMP operator for the purpose of setting access prices should reflect the higher risk of investment relative to investment into current networks based on copper."
Recent results from Virgin Media indicate that take-up of higher bandwidth services is increasing, but it seems there is much that needs to happen to drive investment, from a regulatory as well as demand perspective. The need to drive the marketplace to deliver something it wouldn't necessarily deliver by itself, ahead of current customer demand, or perhaps to deliver it much earlier than it would otherwise, seems to be the Commission's bold goal here.

I don't disagree with the ambition, or the importance of fibre in the longer term, but I do wonder how realistic this is in the light of current funding, roll-out plans and customer demand? This roadmap from BT provides an interesting counterpoint, demonstrating what FTTC services may be capable of in future.

Finally (!), a new guide on broadband investment models, commissioned from Analysys Mason, will be available shortly. Discussed at the first Digital Agenda Assembly in June (see this presentation) and also at the DG Regio European Open Days (see this presentation), the guidance will provide advice on planning broadband investments, identifying five investment models:
  • Bottom-up: group of end users oversee the contract to build and operate their own local network
  • Private design build and operate: managing authority provides a grant to private sector to assist in deployment of new network
  • Public outsourcing: single contract for construction and operation of network, but public sector retains ownership and some control
  • Joint venture: ownership of the network is split between the public and private sector
  • Public design build and operate: public sector owns and operates a network without any private sector assistance
The guide will be illustrated from findings from operational projects across Europe matching these models. It's due to be available by the end of this month.