Development of existing and alternative last mile infrastructure: - Models of Local Loop Unbundling; - Regulatory actions needed for successful applying of LLU positive and negative experience. Natalija Gelvanovska Head of Access and Networks division Communications Regulatory Authority of the Republic of Lithuania (RRT) The view expressed in this presentation is a personal view of the author and do N.Gelvanovska 2010.04.07 1 Slide not necessarily represent the view of RRT
LLU Why? Unbundling, as a policy, is built on the recognition that incumbent carriers have a dominant position in the provision of local communication access by virtue of their control over the local loop, which < > is considered as an essential facility that cannot be economically replicated by alternative operators. Notwithstanding liberalisation of telecommunication markets, it has proven extremely difficult <...> to reduce the bottleneck control of incumbents over the local loop and access to this loop. Source: OECD N.Gelvanovska 2010.04.07 2 Slide
LLU History The history of facilities-based competition, especially in the local access market, proves how difficult it is to compete for customers in this market; Approximately 9 19% of market share in ~16 years (1984(5) 2001): United States: 9% of the local subscriber market; United Kingdom: 19% of the residential lines; Japan: the share of new entrants 18.5%. An alternative approach to LLU, discussed by a few countries, would have been structural separation, putting the ownership of the local loops into a different company. N.Gelvanovska 2010.04.07 3 Slide
What is the goal? LLU In 2000 2001 majority of EU member countries began implementing LLU by requiring incumbent telcos to make their local phone loop available to third-parties; LLU was introduced as a way to overcome the bottleneck control over the last (or first) mile of copper cable owned by incumbent operators which had proved very hard and very expensive to replicate; With LLU alternative operators can gain full control of the network that connects them to their client base, and differentiate their services from those offered by the incumbent; The goal was to allow competition to emerge in the calling market and with introduction of xdsl in the broadband market. With introduction of xdsl, LLU was moved to N.Gelvanovska the forefront of the Regulatory agendas 2010.04.07 4 Slide
LLU and ladder of investments Full Unbundling Line Sharing Bitstream Access Simple Resale Introduction of LLU marks a major break from ISPs simply reselling the incumbents DSL offers, or marketing bitstream offers unbundling also requires a greater investment. N.Gelvanovska 2010.04.07 5 Slide
requires a lot of REGULATION Progress in LLU in the majority of countries has depended on: Progress in achieving consensus on collocation procedures and pricing. The availability of collocation space in some exchanges where a number of operators want space at the same time and procedures to allocate this space. The pace at which regulators have taken action to solve problems in applying LLU. The extent to which rebalancing of subscriber line prices has been undertaken in the past to ensure that there is no price squeeze on new entrants: A major element of success of unbundling is linked to the relationship between the price of the subscriber line to the end-user and the cost of unbundling for the new entrant. Successful LLU implementation requires continuous: regulatory intervention and also Successful LLU co-operation between incumbent and unbundler. N.Gelvanovska 2010.04.07 6 Slide
in EU is mid-2004 Since 2004 LLU has become one of the prime strategic growth paths for internet service providers (ISP) in EU: from fewer than 800,000 lines in the first half of 2004 to more than 2 million in subsequent 6-month periods; The unbundled line base doubled in 2005; Why? Few groups of reasons may be named: Market situation: i.e. the drop in the price of xdsl equipment; Regulatory actions: Take up of LLU i.e. the decrease in rental and connection tariffs and improved procedures following regulator intervention. But not only this N.Gelvanovska 2010.04.07 7 Slide
Take up of LLU, Broadband demand and changing Internet usage The global IP traffic more than doubled during last two years and for the majority of this traffic is responsible consumer video; Massive amount of traffic is consumed by IP traffic of video on demand, Internet video, exchange of video with other users through P2P: In 2008 12 billion video streams were consumed over the Internet each month in US and 95 proc. of that is user generated content; In 2Q 2008 25% of all Skype calls were video calls Source: ebay Internet usage is much less representing classical communication model in a form of client server, i.e. passive downloading of content; Users distribute the content they have created to online social networks and blogs: Increasing worldwide broadband penetration; rocketing sales of digital cameras and mobile phones with photo and video capabilities; e.g. 50% of Korean Internet users report having N.Gelvanovskaa homepage 2010.04.07 and/or a blog 8 Slide [Source: Korean Internet White paper, 2008]
Video component is going to have an impact on broadband environment and firstly on infrastructure. N.Gelvanovska 2010.04.07 9 Slide
Portion of video traffic is liable to be shifted into mobile networks, but The vast majority of video traffic will remain in fixed networks; We observe emerging social aspects of video. Video is becoming an integral part of communication services; The conclusion is that video is definitely going to have an impact in the broadband environment: The access networks are facing challenges of scale So why should we know all this? LLU and Video Changes in Internet usage form new Requirements for broadband access technologies: e.g. not just faster, but also more symmetrical; Increasing Video component within overall traffic pattern is pushing operators to upgrade (replace) their access networks. Changes in the Access networks of incumbents have the N.Gelvanovska 2010.04.07 10 Slide direct impact on LLU
Broadband markets are growing... and Access networks are changing First ALL IP transformation plans: British Telecom's 21CN plan, France Telecom's NEXT strategy, Singapore Telecom's network transform, China Telecom's CN2 and KPN's ALL IP Projects. Movement towards all-ip networks include modernization of access networks: FTTB/C + ADSL2+/VDSL2 become a primary network model while FTTH become a complementary method for network construction; Improvement of network equipment also contributed to faster take up of ALL IP projects, e.g.: introduction of IP DSLAMs (instead of ATM DSLAMs) made it easier provision of Triple-play; new equipment decreases power consumption. N.Gelvanovska 2010.04.07 11 Slide
Broadband markets are growing How Access networks are changing? First deployments (view from 2007) VDSL dominates: Deutsche telecom, Swisscome, Belgacom, KPN, TDC FTTN+VDSL solution; Telecom Italia mixed FTTN+VDSL and FTTB; Only FT selected FTTH (GPON). In more recent deployments portion of fiber increases: TEO LT, AB FTTH (GPON, Active Star) dominates. With introduction of VDSL only sub-loop unbundling became possible. N.Gelvanovska 2010.04.07 12 Slide
NGA and LLU: Subsequences Fibre to the cabinet (FTTC) have significant adverse consequences for local loop unbundlers: particularly if the incumbent were to sell off its existing exchanges, meaning that local loop unbundling (LLU) at the exchange was no longer available. Unbundlers had to choose between withdrawing to a wholesale DSL product and investing in sub-loop unbundling at the street cabinet; The costs of sub-loop unbundling are substantial: Firstly, there is the cost of new equipment at each cabinet, likely to include a multiservice access node (MSAN) to deliver voice and data services, and the establishment of co-location facilities; Secondly, there is the cost of acquiring a backhaul link to each individual street cabinet; Finally, there is the potential cost of writing off the value of equipment and colocation facilities at the local exchange. The biggest difficulty faced by a prospective sub-loop unbundler is the limited economies of scale that it is able to achieve. Source: [4] N.Gelvanovska 2010.04.07 13 Slide
Subsequences: Economies of Scale Observers presume, that the switch to sub-loop unbundling only makes Fig.: Relative economies of scale for sub-loop unbundling and LLU [Source: Analysys Consulting, economic 2007] sense if ARPU can be increased by at least EUR10 per month. Business users may be willing to pay such a price, but will the mass market N.Gelvanovska 2010.04.07 14 Slide really pay this much?
Different FTTx options NGA networks Dwelling Units (SDUs) (e.g. individual residences, houses), Multiple Dwelling Units (MDUs) (e.g. apartments). Source: Alcatel-Lucent N.Gelvanovska 2010.04.07 15 Slide
What to unbundle? New unbundling products Access to in-house wiring Concentration point unbundling Sub-loop unbundling or unbundling on the cabinet Full Unbundling Line Sharing Bitstream Access Simple Resale With development of NGA networks, ladder of investments became more sophisticated. N.Gelvanovska 2010.04.07 16 Slide
LLU success: what are the lessons? The history of successful unbundling proves that: Unbundling has fuelled the rise of competition in the broadband market, particularly in those countries where competition from alternative technologies (e.g. cable) is weak or non-existent; France is a prime example here, with 94% (in 2006) of broadband connections based on DSL: incumbent's share of the broadband retail market dropped from 70% in 2002 to 47% in 2005; Success of unbundling directly depends on the ability to reach economies of scale. Number of preconditions may be determinant: Incumbent s network structure, in particular the size of the exchanges; Proportion of multi-tenant buildings, etc.; In case preconditions are not ensured, regulation may have limited impact on LLU take up; In case preconditions are favorable, regulation is next determinant factor for successful LLU development. N.Gelvanovska 2010.04.07 17 Slide
LLU success: what are the lessons? Is poor LLU take up always a Negative experience? Last upper steps of the ladder of investments take alternative operators too close to the decision to develop their own infrastructure; Access to in-house wiring Concentration point unbundling Sub-loop unbundling or unbundling on the cabinet Therefore even in case adequate regulation is in place, LLU may not become market choice and alternative operators may decide to BE MORE INDEPENDENT and develop their own infrastructure but isn t is something Regulators want to achieve? N.Gelvanovska 2010.04.07 18 Slide
LLU success: what problems it solves? Economics: Urban vs. Rural N.Gelvanovska 2010.04.07 19 Slide
Broader view on Broadband is the key Broadband infrastructure physically How much of Broadband infrastructure your country has in the ground? National interest is to have as much as possible Right time for discussions Networks are both expanding and upgrading: Expansion of Electronic Communication Services; Natural evolution of infrastructure: In terms of technology: - PSTN NGN; - xdsl technologies will reach their limits during the next decade; In terms of media: Copper wire Fiber optic cable N.Gelvanovska 2010.04.07 20 Slide
Broader view on Broadband is the key Perspective Lifetime of the fiber plant is ~40 years: without any major modification of the passive infrastructure and no introduction of additional components in the fiber plant, it s easy to upgrade transmission speeds... even on a per-customer basis. Wolfgang Fischer, Cisco Systems. Passive infrastructure created today will remain unchanged for 20 40 years N.Gelvanovska 2010.04.07 21 Slide
Broader view on Broadband is the key What is difficult about Broadband development To develop a physical infrastructure is rather very expensive NGA deployments: FTTN in Germany for 37% of population: 4,8 billion ; FTTH in France for 7% of population: 3,4 billion ; FTTN: Deliver minimum downlink speeds of 12 Mbps to 98% of Australian homes - 4.7 billion AUD to enable the roll-out. Source: WIK Consult Rather billions than millions of Investments N.Gelvanovska 2010.04.07 22 Slide Source: IDATE
Broader view on Broadband is the key In more details Civil works represent 60 90% of the total rollout costs: 80% of the total rollout costs of NGA Source: European Commission Cost consideration for P2P Ethernet Access network: N.Gelvanovska 2010.04.07 23 Slide Source: Corning & FTTH Council Europe
Broader view on Broadband is the key Who will own tomorrow s infrastructure? other sectors (i.e. energy) participate in construction of Passive infrastructure suitable for Ultra-Broadband provision: N.Gelvanovska 2010.04.07 24 Slide
Broader view: Lithuanian case Symmetric obligations: Symmetrical facility sharing obligation under the Law of Electronic Communications in place since 2004: Detailed regulation in place since 2005; Reference offer for Duct access available from 2005: Successful commercial product; Significant number of networks were developed (i.e. cable); Dispute resolution procedure adopted to resolve facility sharing disputes; Amendments for guidelines of construction of electronic communications networks extended to explicitly cover infrastructure construction in blocks (of flats): focusing on inlets to the buildings, ducting and in-building wiring. N.Gelvanovska 2010.04.07 25 Slide
Broader view: Lithuanian case Asymmetric obligations: Obligations in Market of Unbundled access to the local loop remain in place: No real interest in LLU Mainly due to reasonable conditions for Access to ducts; New market of access to physical infrastructure (new market 4) is under research final decision planned in autumn 2010. Number of LLU is just over 500 lines, but at the same time Lithuania is the economy with the highest penetration of the FTTx in EU. N.Gelvanovska 2010.04.07 26 Slide
Key findings & Conclusions Successful LLU implementation requires continuous regulatory intervention because of the imbalance in negotiating power between the incumbents and new entrants; LLU requires constant operational co-ordination between the incumbent and unbundler: ordering, provisioning, billing, fault handling and service-level agreements, pricing, collocation and spectrum management, etc. Nevertheless successfully implemented, LLU alone is not sufficient to address all the problems in the broadband market and deployment of the alternative technologies remains of key importance. N.Gelvanovska 2010.04.07 27 Slide
Key findings & Conclusions In the context of the NGA networks LLU: Become more sophisticated: unbundling from the cabinet, unbundling from the basement of the building, etc.; Particularly requires regulatory attention to vertically integrated associated facilities, i.e. ducts, dark fiber, etc. Successful LLU requires numerous preconditions, i.e. network structure, which are may be not related to regulatory frameworks; LLU should not become an objective itself, the objective is competitive broadband market: LLU is just one of the tools form the regulatory toolkit which should be used where necessary: Application of LLU is normally limited to one undertaken and may have economic sense mainly in urban areas. N.Gelvanovska 2010.04.07 28 Slide
Key findings & Conclusions Both natural technological changes and expansion of electronic communications determine that networks are upgrading and expanding: Civil works are clearly representing an economical bottleneck for development of access networks; policy decision could reduce roll-out costs and contribute to further broadband development. Created network infrastructure will remain unchanged for 20 40 years; SMP doctrine itself is not enough Possibility to impose Access to physical infrastructure under SMP doctrine is limited in terms of number and type of undertakings; Kind of symmetric obligations should be considered as: owners of fiber access networks are players of different domain; In order to reduce roll-out costs: optimal (re) use of all existing infrastructure should be possible. N.Gelvanovska 2010.04.07 29 Slide
Natalija Gelvanovska ngelvanovska@rrt.lt Communications Regulatory Authority of the Republic of Lithuania www.rrt.lt N.Gelvanovska 2010.04.07 30 Slide
References: 1. Unbundling in Europe: Recent Trends by Sophie BISMUT, idate, Communications & Strategies, no. 63, 3rd quarter 2006, p. 167., at: http://www.idate.org/fic/revue_telech/171/cs63_bismut.pdf; 2. IP DSLAM: Pioneering the IP convergence era, Huawei, at: http://www.huawei.com/articles_papers2/archive/simpleres.do?id=1525&type=technology; 3. Cost analysis for DSL networks, ICT Regulation Toolkit, ITU, at: http://www.ictregulationtoolkit.org/en/practicenote.aspx?id=2899; 4. Is fibre to the cabinet a disaster for local loop unbundlers?, Catherine Hammond, Analysys Consulting, February 2007, at: http://www.totaltele.com/view.aspx?c=0&id=349031; 5. DT Revamp Puts FTTH in Focus, Anne Morris, Freelance Editor, March 17, 2010, at: http://www.lightreading.com/document.asp?doc_id=189297; 6. Policy principles for the Irish NGN regulatory framework, Ovum, 2007, available at: http://www.comreg.ie/_fileupload/publications/ngn_ovum_report_2007.pdf; N.Gelvanovska 2010.04.07 31 Slide