Earlier this year, I attended a dinner for about 150 people. The first thing we all did was head for the bar to grab a drink, swamping the staff and setting back the dinner schedule by nearly two hours. We learned that it is better to place a bottle of cheap wine and a jug of lemon squash on each table just to manage the crowd. The same lesson is now being applied to the NBN.
Last Sunday saw the long-awaited signing of two deals and new rules which will, hopefully, pave the way for the National Broadband Network under the current government’s Multi-Technology-Mix model. It’s a positive development for getting more people reasonably happy sooner, but there are significant long-term challenges ahead.
The Carrier Licence Conditions (Networks supplying Superfast Carriage Services to Residential Customers) Declaration 2014 is intended to manage so-called cherry-picking of profitable residential customers of fast broadband. The previous model under the Labor government discouraged anyone other than NBN Co from rolling out the local access network. In part this was to break the vertical integration stranglehold Telstra has over both retail and wholesale provision of access through its legacy plain old copper network; it was also to ensure NBN Co would not be undermined by competition. However, a loophole in the regulations was identified by TPG, allowing it to reach up to 5% of the NBN footprint as a retail provider of Fibre-to-the-Basement through its own network.
The new regulations recognise the advantages of other parties choosing to roll out their own networks, but require that such networks be available on a wholesale basis to other retailers – which is to say there is a requirement to structurally separate wholesale and retail operations on NBN-equivalent services. The regulations only apply to new networks rolled out after 1 January 2015. Services already in use, such as iiNet’s TransACT fibre-to-the-node network are excluded. The wording has clearly been carefully sculpted to capture TPG’s plans.
Deals with both Optus and Telstra were also signed. Both companies have agreed to progressively hand over ownership of their relevant infrastructure, Telstra’s copper access network and the HFC cable networks of both, to NBN Co. Previously, the arrangement was that these networks would be decommissioned, but they are now critical implementation options for the NBN. Agreements are consistent with maintaining the valuations of the previous arrangements so that shareholders are no worse off.
Mention has also been made of actively tapping Telstra’s experience and capabilities in the NBN rollout. There is some sensitivity around this statement from the Department of Communications, after Telstra was effectively excluded from tendering to build the NBN in 2008. There would appear to be a significant mending of relations.
But there are troublesome times ahead. The original Fibre-to-the-Premises-dominated NBN was big on vision but caused an overnight drop in investment in moderate-speed broadband technologies such as ADSL. This meant customers were stuck with 2008-vintage speeds while NBN Co geared up for planning and deployment. The political fix negotiated by independents Tony Windsor and Rob Oakeshott in 2010, which shifted priorities to rural areas, provided some longer term certainty for infrastructure investment in profitable urban areas, but further delayed urban rollout. The new policy settings encourage infrastructure competition on the basis of structural separation, once again providing opportunities for parties other than NBN Co to compete in the wholesale space.
However, the new NBN model is focused on initial speeds of 25MBit/s up to 50MBit/s in the short term, but with no mention of Committed Information Rate or specifics of upload speeds. This myopic vision fails to recognise that the NBN is not just about “more of the same, but faster”. It is entirely about enabling fundamentally new services to domestic households, both for entertainment and for productive work-at-home.
What we’re left with
The emerging NBN is barely capable of delivering a couple of high-definition video streams in real time, yet this is common in a modern family household with independent internet-enabled devices. If you want to download a one-hour standard definition video, it will take about five minutes. However, if you want to produce and upload such video content, it will take three hours. Upload figures for high definition and the emerging ultra-high definition video are about fifteen and thirty hours, respectively. Cloud-based services will continue to stumble as long as policy direction fails to recognise the importance of delivery of data into the network.
The much deeper issue is what happens after 2016. I’ve argued previously that “[o]ptical fibre is the only known viable technology beyond 2025. The only justification for considering anything else in the meantime is to buy us time”.
A Fibre-to-the-Node deployment will at least provide the option of upgrading to a passive optical link to the premises of customers who want it, and some of us just can’t wait. But there is no guarantee that every other technology deployment has a similar evolution option. It would be wise for competing infrastructure providers to think this through, but there is no guarantee that solutions deployed in the next few years will evolve, efficiently, to the robust projected demands beyond 2025.
So is this good news? Actually, yes. The NBN capital expenditure won’t be any cheaper if we consider a realistic 20-30 year timeframe, but that doesn’t matter if the faster deployment of the interim Multi-Technology-Mix generates revenues and productivity gains sooner, offsetting peak funding. The interim solutions won’t deliver a clean evolutionary pathway, but that doesn’t matter if enough customers are sufficiently satisfied in the short term. The long-term fibre-to-the-premises deployment can be managed more calmly, if less efficiently, over a longer time frame.
The line at the bar is still too long. Pass me the bottle, please.
This article originally appeared at The Conversation.