Robin Eckermann, who led the creation of TransACT and served as its chief architect during the 2000-2003 network rollout, said on Monday that the government's creation of NBN Co to lead the implementation of upgraded network infrastructure could, in some ways, be seen as an admission that the strategy implemented in the 1990s had failed.
He was asked to look at the suggestions made in these columns last Friday by Dr Mark Gregory, an associate professor in network engineering at RMIT University, regarding the future of NBN Co after the rollout of the broadband network is over, and offer his opinion.
Eckermann indicated he had a sense of déjà vu when looking at what NBN Co was becoming: "NBN Co is evolving to have much the same dominance that Telstra had historically." And, he added, "One can imagine a skit by the late John Clarke and his partner Bryan Dawe: 'Minister, we've found a way of breaking Telstra's monopoly!' 'And how are we going to achieve that?' 'We're going to build a bigger and uglier monopoly to take its place'!"
"I think many Australians are concerned about the principle of putting monopolies into private ownership," Eckermann said. "Most would see the track record of privatisation in the electricity sector as little short of a disaster - and I think they would have grave concerns about the prospect of their only fixed broadband connection transferring into private ownership where maximising shareholder returns is a key objective."
On the goal of increasing average revenue per use to $52 — the level that NBN Co says it will need to break even — he said it was both problematic and naive. "I commonly hear the complaint that with NBN Co, costs have gone up and performance has gone down. In many cases the reasons for this have little to do with the last-mile technology and all to do with congestion levels - in CVC capacity and/or backhaul," he added.
Eckermann said if people had to make a choice about telecommunication costs, "I expect that many will conclude that they cannot get by without their mobile services - so it's fixed broadband that will have to go".
"Already we have seen the mobile network operators bring out plans with broadband allowances that are quite adequate for the many users for whom streaming video services and gaming aren't priorities," he added.
"The potential of the mobile networks networks to offer an alternative to fixed broadband will increase when 5G begins to be widely deployed from about 2020. In the light of this, any success in increasing ARPU is likely to be offset by declining connection rates, resulting in little if any gain in overall revenue."
He cited what he said was "one of the most astute observations" about the NBN Co which be attributed to well-known entrepreneur Bevan Slattery: that one could have any two of performance, affordability and return on investment.
"If Australia does not prioritise the first two of these, any vision for the NBN to be the foundation for Australia's place in the increasingly digital world of the future becomes a joke!" he said.
Eckermann pointed out that the history of access network investments in Australia was littered with write-downs of the original network cost - "witness, for example, the HFC networks of Telstra and Optus and TransACT's FttK/VDSL network. Do we have reason to believe the NBN will escape a similar fate? With a reduction in the written down value of the network, the nationally important objectives of affordability and performance could be achieved."
He suggested that businesses should be charged more than individual users, as they would be prepared to pay significantly more for good telecommunications services. "And I have little doubt that NBN Co has good potential to improve revenues in this market segment. However, if residential ARPUs rise much above present levels, we will have a serious problem!"
Eckermann was of the opinion that while the NBN would evolve technologically over time, the goal of achieving 93% fibre-to-the-premises would never return.
"Fibre-to-the-kerb (kerb is the Australian spelling, though NBN Co likes to use the American spelling 'curb' and the acronym FttC or FttDP) is capable of performance into the hundreds of Mbps with even higher performance being demonstrated in the laboratories," he said.
"In the early life of the post-deployment network, I expect upgrades to focus on extending the fixed wireless footprint (to alleviate pressure on satellite capacity), decreasing the number of users per fixed wireless sector (to improve fixed wireless performance) and upgrading FttN areas to FttDP (where long copper lengths are unable to deliver the level of performance people want (and) for which they are prepared to pay).
"I personally view the thought that wholesale competition in a privatised NBN world might inspire a massive new spending spree to upgrade the bulk of the network to FttP as being naive and fanciful. However, as FttK proliferates, the migration of individual customer circuits to FttP for the handful of users requiring more than what FttK can deliver is a comparatively minor upgrade."