This week we learnt that tests by NBN Co have achieved broadband speeds of up to 8.3 Gbps over short runs of copper. This is significant given the length of these copper runs are of a similar (or greater) length than is needed to deliver the NBN to consumers using fibre-to-the-distribution-point (FTTdp) technology. Obviously these are experimental results. They show the theoretical capability of such technologies although their practical deployment may be a considerable way off.
The irony of this announcement is that Australian consumers are currently being denied the optimum broadband speeds that are already technologically and practically achievable today. It is almost impossible to believe that the broadband speeds available over the NBN are being artificially restricted, not by the technological capabilities of the network, but by a pricing model that denies end users the very benefit that the NBN project was designed to deliver.
At the outset of the NBN project there was a lot of public discussion around guaranteed broadband speeds of 100Mbps, and whether in fact such speeds were necessary, or whether a much more modest figure of perhaps 25Mbps or even 12Mbps would meet Australian’s needs for the foreseeable future. More recently we have experienced the acrimonious debate around the so-called multi-technology mix (MTM) and whether this provides an adequate alternative to the original fibre- to-the-premises (FTTP) model.
All of this discussion centred on so called “last mile” access technologies. That is, what medium was used to connect your house to the network; was it to be a new optic fibre cable; would the old copper wire be used; or perhaps the already connected pay TV cable? In NBN terminology, this connection is referred to as the access virtual circuit or AVC. However, whatever is used to connect to your house, that circuit needs to be connected to the network, aggregated with others, and carried back to the point of interconnection, where retail service providers (RSPs) can connect to the NBN and provide connectivity to their own and other networks and to the internet. The part of the network that connects the AVC from each house and links them all back to the point of interconnection is referred to as the connectivity virtual circuit (or CVC) (see diagram below). It is the pricing of this component which is currently undermining the efficiency of the NBN.
Click here to view graph.
The wholesale cost of an AVC capable of providing 50 Mbps downstream and 20 Mbps upstream from NBN is about $34 per month. However, in order to provide your service, the RSP is also forced to buy CVC capacity at a price of between $15 and $20 per Mbps per month. This cost is so high that it is effectively impossible for RSPs to offer an adequate broadband speed to consumers at a reasonable price. This is because, at busy times, when a high proportion of people are using the network, regardless of the speed and capacity of the AVC connection to your home, you will not be able to receive a broadband speed any higher than the amount of CVC capacity that your RSP has provisioned. Currently RSPs have provisioned on average only about 1 Mbps per subscriber on the NBN. So, at peak times, rather than receiving the 12, 25, 50 or 100 Mbps service that their AVC connection is capable of delivering, they are actually receiving a broadband speed of approximately 1 to 2 Mbps. Hardly the broadband nirvana we were promised, and clearly inadequate to meet the requirements of a digital future.
On the current CVC pricing model it would cost approximately $210 per month in CVC costs alone for an RSP to procure sufficient CVC capacity to be confident of providing an actual busy hour connection speed of a relatively modest 12 Mbps. Once other input costs are taken into account, it is unlikely to be possible for an RSP to provide such a service on a retail basis below $300 per month. This is well out of reach for most consumers, and much higher than international benchmarks for comparable services.
So, while it is encouraging to hear of the theoretical possibility of future data speeds of 8.3 Gbps, at a CVC cost of about $150,000 per month to be confident of receiving that speed when you really want it, it is unlikely to be within reach for most of us. And while NBN Co has flagged the possibility of reducing CVC pricing to as low as $10 per Mpbs per month next year, a price tag of $83,000 per month for CVC capacity still seems a little excessive! In fact, even at that discounted price, a busy hour speed of only 25 Mbps would come in at $250 a month for CVC capacity alone.
The primary objective of the NBN project was to provide abundant, affordable broadband capacity to the whole of the Australian population, to boost productivity encourage innovation and provide Australia with the competitive advantage of state-of-the-art communications infrastructure as it heads into a digital future. The NBN is now being built by the Government, using public money, at vast expense and on the back of unprecedented political and regulatory turmoil and upheaval. It will be a hugely disappointing outcome if, having made the investment, we allow it to founder on an inefficient and retrograde pricing structure which creates an artificial scarcity of the very commodity the project was intended to supply.