commentary Now that the board of the new National Broadband Network company, NBN Co, has been assembled, the moment of truth is nearing.
There is little prospect that the NBN Co can be financially sustainable on a stand-alone basis, which means the original concept of a giant public/private partnership will founder without massive and ongoing government subsidies.
There is little prospect that the NBN Co can be financially sustainable on a stand-alone basis
The telecommunications team at Goldman Sachs JBWere has just released a major 92-page report on the NBN. It estimates it will cost $37 billion to build — $41 billion if Telstra isn't prepared to sell the ducts, pits and pipe that constitute the most strategic element of its "last mile" network to NBN at a 33 per cent discount to the analysts' $12 billion valuation — but be worth negative $9 billion in net present value terms.
As the team concludes: "It is difficult to see the market ascribing any value to an equity investment in a company such as this." It forecasts the NBN Co won't be cash flow positive until 2025.
The team argues that Telstra will be prepared to sell its passive infrastructure to the NBN Co for $8 billion — a $4 billion discount to its assessed value — to demonstrate that it is a good corporate citizen (and presumably to try to avoid regulatory punishment for non-cooperation).
Telstra would also significantly reduce its maintenance capital expenditures and selling its assets to NBN Co would avert the threat that the NBN would "go aerial" and, in the longer term, leave Telstra's existing network intact but eventually obsolete.
However, the Goldman analysts don't believe Telstra will accept equity in NBN Co as consideration, given their view of its equity value. They believe it will be politically unpalatable for the government to pay cash for the assets, saying the most likely outcome was a mixture of cash, the transfer of some Telstra debt, and some kind of annuity stream.
The most interesting/controversial aspect of the research is the team's modelling of the NBN Co. They say it will be 2017 before 50 per cent of Australian homes are passed by the network, and 2025 before the network passes 90 per cent of Australian homes. By 2020 40 per cent of the homes would be connected and it would take until 2028 for 85 per cent of homes to be plugged into the NBN.
That means the NBN Co would be burning a lot of cash for a long time — more than $2 billion in 2012, with negative cash flows continuing until about 2025. At an earnings before interest and tax level, it would take until 2019 for the network to break even before moving into positive territory, with Goldman estimating EBIT of $425 million in 2020.
While the team retains doubt that the NBN will be built, that will ultimately be a matter of the government's resolve and preparedness to write a blank cheque in the form of funding and funding guarantees. If it is right, however, it will be very difficult for existing telcos to vend in their own assets in exchange for equity, a key concept of the original proposal, if there is no equity value in NBN Co.
The analysts believe that Telstra will retain its copper network, selling and leasing back access to its ducts and pits, which would mean the NBN's fibre-to-the-home network would face competition from the cheaper copper network and the take-up will be slower than if the copper were also acquired and shut down.
Goldman believes the government would be reluctant to pay for the copper network, which will ultimately be obsolete, and would also be wary of a backlash from consumers if they are forcibly migrated onto a higher cost platform. Acquiring the copper and continuing to operate services on it would increase the cost of acquiring Telstra's infrastructure to $20 billion.
Acquiring the ducts, pits and pipes at a material discount would not only reduce the overall build costs but would, the analysts say, cut the cost to pass/connect in metro areas from $3500 per home to $2500 — about $12 billion in total — and reduce the time frame for the roll-out by three or four years.
There could also, the team says, be a prospect of the government having to pay compensation to internet service providers for their redundant digital subscriber line access multiplexers if a forced migration were pursued.
They also argue that it is probable the NBN Co is unlikely to simply provide dark fibre services (the least sophisticated and cheapest model) or to provide a full suite of wholesale services (the most sophisticated and expensive model). Instead it was probable NBN Co would provide bitstream services, selling data services to wholesale service providers that would then develop and on-sell products and services to retail service providers.
The "two-layer" nature of that model would, they argue, mute the competitive impact of the NBN on Telstra in the short term because there would need to be significant investment at the wholesale level before there could be significant competition at the retail level. Telstra could also compete with other wholesale service providers to win the business of the retail service providers.
It is clear that the new NBN board is going to be faced with a myriad of different choices, in terms of the nature of the network, the services it provides, the way it is rolled out and funded and how Telstra and other telcos and the new network interact and cooperate.
The Goldman report reinforces the market's conclusion that, whatever the NBN looks like, it is going to have to be taxpayer funded and the cheques will be massive.
This article by Business Spectator's Stephen Bartholomeusz is reproduced on ZDNet.com.au courtesy of a reciprocal publishing agreement.
4%
1%
We all have to sit and wonder at the progress of the NBN. I do not live in a country area, Safety Beach being, could I say, an outer suburb. I am at the moment using Telstra 3G with better results than people able to receive ADSL in the close vicinity. That is within two hundred metres. Surely there is a good reason to look at wireless as a less expensive way of supplying the Internet to the country regions. At the moment 7200 Kbts is achievable in the city areas and the outlook is for faster speeds in the future. If wireless was availiable to more people at this speed then the uptake would be greater and the plan prices may become lower because of volume of users. Is it cheaper to build a tower than bury cables in the ground?. Surely it has to be