Overnight, industry pundits measured their celebrations over the Federal Government's plans to build its own national broadband network with attempts to calculate how much such a network might cost for the average consumer to access.
The Government expects the public/private company that builds the National Broadband Network to spend some $43 billion in the process.
But no investor is going to front up for Aussie Infrastructure Bonds without some assurance the company will make a handsome profit in the long-term.
The resulting price of NBN services, said iiNet managing director Michael Malone, "is the real question."
"If the access price is too high, the new network won't compete with [mobile] 3G, ADSL [copper fixed line] or HFC [cable]."
Scenario 1: Assuming full-take up
The analysis thus far suggests that the wholesale cost of a service will have to be at least $50 per month just to cover the cost of building the network.
Internode Carrier Relations manager John Lindsay and independent blogger Duncan Reilly both came up with this figure, through slightly similar means.
Both rely on Australian Bureau of Statistics projections that suggest Australia will have eight million households by 2009, equating to a network cost of roughly $5,000 per user assuming that all Australian households take up the service.
Assuming that the network is paid off over a ten-year life (120 months) Lindsay projected the wholesale cost of the network is $50 per user per month.
From here the guesswork gets tricky. How much margin does a wholesaler expect? How much margin will retailers put on top?
Current ADSL2+ prices are a good guide.
According to Lindsay, the cost of a wholesale ADSL port is between $24 (for 256K) and $60 (for ADSL2+) per month.
"It isn't obvious, but the cost to the wholesale provider per port is actually fixed and is around $30," Lindsay explained. "They sell entry level at break even and take a great profit at the highest speed."
Linking these ports back to capital city points of interconnection cost $120 per megabit, he said, while connection to the global Internet costs roughly $200 per megabit.
One can assume then, that today's retail ISP is paying a fixed cost of between $24 and $60 per service, plus a variable cost of $1 per gigabyte on their network.
At $50 per service, Lindsay suggested that the projected NBN would cost a retail ISP $20 more than the underlying cost of an ADSL port today, but that this "might well balance out with telephony and other perhaps non-Internet services."
In this ideal scenario, prices could be much the same as today, "tending lower with a bundle," he said.
This calculation, however, assumes that the network is fully utilised.
"What if the network only gets two million users?" Lindsay said. "What if the rest stay on ADSL and wireless?"
Scenario 2: Assuming limited take-up
The only consumers of high speed broadband you can truly bank on to take up the NBN, he said, are the two million users prepared to pay for services of 8 Mbps or higher today.
"So now we need to spread the cost over four times fewer users, which means the per-user cost is four times higher," he said.
Suddenly, the opposition's claim of the network costing consumers $200 per month sounds a little more feasible.
"Replace the $60 per month cost of an ADSL2+ wholesale port with $200 and your retail price has to increase by $140 before you break even," Lindsay said.
Scenario 3: Finding a middle ground
Even if user demand fell somewhere between the two extremes - at say, five million users, retailers are still looking at a wholesale cost of $70 per month.
And if the National Broadband Network Corporation (whatever form it takes) only expected five per cent profit margins, and retailers expected 10 per cent, consumers should expect to pay upwards of $85 per month for their FttH connection.
Today, for example, Internode offers 100Mbps FttH connections at between $99 and $185 per month, with generous download caps included.
A price of $85 per month would not seem unreasonable for a regional user that currently pays upwards of $70 and $100 per month for ADSL2+ coupled with a reasonable download quota.
The question for metro users, who pay only $40-$60 for very good quality connections, is whether they would pay twice that price for a connection that offers four or five times the speeds.
They would, perhaps, if it meant avoiding Telstra's line rental fees of $25-$30 per month.
They would also be prepared to pay a premium, said Hostworks sales manager Klaus Bartosch, if the applications pushed down the pipe carry new, as yet unconceived value for consumers.
"Younger generations might pay if the content is at the right quality and they are getting new value from that service," he said. "In that case, price won't have much bearing on their decision."
External factors
David Forman, executive director of the Competitive Carriers Coalition believed that comparing prices from the new network to our current wholesale costs will be irrelevant in eight years time, considering the Government's proposed changes to telecommunications regulation.
"We will be looking at this through an entirely different prism," he said. "Regulatory reforms will in the meantime drive prices down."
Peter Coroneos, director of the Internet Industry Association (IIA), agrees.
"I think we can infer that forcing contestibility in wholesale access must result in lower prices as a natural consequence of levelling the playing field between retail access providers," he told iTnews.
Similarly, Lindsay expected the premium paid by business users to access the NBN will subsidise retail plans.
"The thing that will influence the price more than anything is the premium paid by business users for high availability and symmetric services," he said.
"If business customers cost the retailer $500 per month and make up 10 per cent of the users, the network operator has $250 million of monthly revenue before having to recover the balance of the $350 million needed to cover costs from residential," Lindsay calculated.
"That would result in a $23 cost to retail ISPs for residential. That's a great price."
Conclusion
"The Rudd Shiny New Network is courageous," Lindsay concluded. "It will work technically and it's definitely the best solution. It might work financially if it's done right, particularly getting the price balances between business versus residential and metro versus regional right.
"I think the best way to ensure financially viability is to start where the demand is highest: the metro black-spots and regional centres. You get the highest take-up there. Then work towards the CBDs.
"That is the nation building option."
-1.jpg&w=100&c=1&s=0)
.png&w=100&c=1&s=0)
