If the ACCC follows the line presented in a draft determination issued yesterday, then it will add Ethernet services to its list of declared “Domestic Transmission Capacity Services”.
Such a declaration would come fairly quickly – the competition regulator is accepting submissions on its draft report until August 20, after which it will make its final decision – so now is a good time to explain why the decision will matter to Australian businesses.
Over the past decade, Ethernet has well and truly broken its original boundaries as a LAN protocol. And in the last five years, it has become a serious contender in the wide area networking market – both within the telecommunications industry (for interconnection between carriers) and outside it (for customer connections).
A lot of users of Ethernet telecommunications services may not even know they’re using an Ethernet service: for example, while ATM was for many years the preferred transport layer connecting ADSL modems to DSLAMs, Ethernet has been supplanting it for some time now. Ethernet – whether delivered on fibre for short distances or over copper (at lower speeds) for longer distances – has also become highly popular in the business market.
But it’s the growth of Ethernet in the telecommunications carrier market that has brought it to the attention of the ACCC.
The growth of the competitive telecommunications market creates a need for carriers to interconnect to each others’ networks, and in particular, to Telstra’s network. Like it or not, Telstra is still the only company that can boast a ubiquitous national network that can deliver a fixed connection to nearly every household and business in Australia (with the exception of a small number of people or companies so remote that they need to use satellite or point-to-point radio services).
Many competitors have deployed their own network infrastructure, particularly in capital cities (while walking near Sydney’s RPA Hospital recently, I spotted pit-covers in the street identifying fibre from four or five different carriers). Where they can connect a customer to their own infrastructure, they don’t need Telstra.
However, any competitive carrier serving customers outside their own footprint will need to buy connection from somebody, which will probably make them a customer of Telstra wholesale.
If the customer is outside that relatively small part of Australia where there is strong telecommunications competition, Telstra is likely to be the only carrier near to the customer site.
So it’s the combination of Ethernet’s market popularity and Telstra’s geographical ubiquity that has made the ACCC consider making Ethernet a “declared service”.
Why declare a service?
The ACCC creates a service declaration when it believes a particular service is necessary to competition.
The ULL – unbundled local loop – service is a great example. The only economic way ISPs could connect their competitive DSLAMs (DSL Access Multiplexer) to customers is to use Telstra’s copper.
Without access to that copper, we would not have a competitive ADSL market, so many years ago, the ACCC designated the ULL as a “declared service”. The practical effect of this is twofold: the ACCC mandates that the service must be made available to Telstra’s competitors; and it regulates key conditions of the sale of the service.
The outcome isn’t always perfect. For example, an ISP may be guaranteed access to the copper in a particular area – but find that it can’t install a DSLAM because there’s no rack space available in the local exchange – but service declaration has been a lynchpin of the competitive market.
A great many services are declared by the ACCC (and since its role extends far beyond telecommunications, it’s worth noting that the ACCC regulates access to far more than just telecommunications infrastructure) – and one of these is the Domestic Transmission Capacity Service, or DTCS.
The importance of the DTCS
Like ULL declaration, the DTCS declaration is designed to promote competition in telecommunications services.
The problem for an ISP is this: once they’ve connected a customer to their DSLAM, the connection has to go somewhere. If the customer is in a hotly-contested city market, that’s not a problem. A number of fibre owners will be competing to carry the traffic, and prices are kept low.
If, however, the DSLAM is in a regional town served only by Telstra, the ISP will have no choice but to buy its backhaul from Telstra. If there were no service declaration in place, Telstra may simply refuse to supply the service, and retain the whole market to itself.
Hence the long-standing declaration covering the DTCS. Although the price of a long-distance backhaul service is set by commercial negotiations, ISPs can at least guarantee that they can source the connection from Telstra.
The ACCC’s declaration specifies the types of service that Telstra must supply, and includes a longish list of regions for which the ACCC considers the service necessary for competition. And while the declaration covers more than just backhaul services, it is backhaul that makes the greatest difference to the competitive market.
Which brings us back to the first question: Why Ethernet?
The current declaration has its feet firmly planted in history. It stipulates a number of speeds – from the relatively lame 2.048 Mbps right up to 155 Mbps – all of which are drawn from the venerable PDH-based and SDH-based services (the acronyms stand for Plesiosynchronous Digital Hierarchy and Synchronous Digital Hierarchy, and that’s enough of them for now).
In other words, today’s service declaration is tied to a “legacy” protocol, and an access seeker with Ethernet “everywhere else” in its network would still need to deal with other types of service wherever the connection is supplied by Telstra.
So the main outcome of adding Ethernet to the list of declared services would be to simplify things for access seekers. While they’re not required to use Ethernet, if that’s the dominant protocol in their network, they can require Telstra Wholesale to supply that service wherever the service declaration applies.
And this would be a good thing for customers, who should enjoy the flow-on benefits of simpler, more modern networks that can be administered with a smaller (and hopefully less-demanding) mix of technologies and skills.