commentary The credibility of the economics of the Rudd Government's proposed $43 billion new National Broadband Network will be tested over the next few weeks as analysts reverse engineer the headline cost to test the business cases that might support it.
The early signs aren't that promising.
The early signs aren't that promising. The maths for the new NBN aren't that complicated and the early analysis suggests that to achieve even modest returns for the government and any private investment it can attract, the new NBN would have to charge access prices to the wholesale network that would result in very substantial increases in retail charges.
The $43 billion estimated cost of the new network (no explanation has yet been proffered as to how that number was arrived at) is to be funded broadly half by debt and half by equity. The borrowings would cost around 7 per cent and, if the investors wanted a commercial return, the equity would probably cost something in the mid-to-high teens for an investment that will carry significant risk.
That produces a blended pre-tax cost of capital in the low teens, or pre-tax earnings of around $5 billion a year. If the NBN could capture all the 10.5 million fixed lines currently in service, that would mean it would need to generate roughly $475 million of pre-tax profits, or about $40 a month, from each line. In turn, that means it would need to charge retailers something significantly more than $40 a month to recover its operating costs and make an acceptable profit.
The early analysis seems to be focusing on wholesale charges of around $45 a month. Today the bigger resellers of Telstra's unconditioned local loop (ULL) generate just under $80 a month and enjoy gross margins of more than 60 per cent. If they wanted to maintain those levels of margins by retailing access to the NBN, they would need to charge something over $100 a month. To maintain prices at current levels they would need to accept a near-halving of their margins.
That is, perhaps, a best-case scenario for the NBN because it assumes the network will displace the 10.5 million lines now in service.
With Telstra retaining its copper network, and perhaps its upgraded HFC cable network in the capital cities, and the economics of ULL access compelling for those telcos that have invested in Digital Subscriber Line Access Multiplexers (DSLAMs) in Telstra's exchanges being very attractive, that may be overly optimistic.
Goldman Sachs' telecommunications analysts have written previously about a dramatic shift occurring within the broadband market, with customers drifting from fixed-line broadband to wireless broadband at an accelerating rate.
The Goldman team estimates that, by 2020, there will be only 8 million fixed access lines in service because of that shift in consumer preferences. Based on the team's numbers, it would require monthly wholesale access prices of about $100 a month for the NBN to deliver a 10 per cent return on the total capital invested in it.
Then there's the retail margin and perhaps, charges for content and pretty soon you're talking a level of cost that could put the network beyond the reach of many households.
Will consumers — and the NBN will need a mass end-market as well as high value business end-users if it is to be economic — be prepared to pay a big premium over their existing cost of broadband access for the 100Mbps speeds? What if Telstra continues to drop, not just the cost of fixed-line broadband from current levels but also reduces wireless broadband charges? What if the ISPs stay with the lower-cost copper network?
The original NBN plan, with a fibre-to-the-node network, would have cut over Telstra's copper lines and therefore removed its network as a competitor. The new NBN plan leaves the copper lines in place and therefore will face a slower but lower-cost rival.
If Telstra holds onto its HFC cable (the possibility of forced divestiture has been raised in the government's discussion paper for regulatory reforms), its cable and Optus' HFC cable, would also be competing networks in the capital cities. Telstra is upgrading its cable to offer comparable peak speeds to the NBN. It has also been musing openly about upgrading its Next G wireless network to speeds of 84Mbps.
The NBN will also offer Telstra, for the first time, the opportunity to reshape its portfolio of fixed-line customers, in effect cherry-picking the higher-value customers and encouraging the rest to shift to the NBN.
The time it will take to build and ramp up the NBN — the government says eight years — and the option of Telstra, and other telcos with their own infrastructure, will have to respond to the new network which means there are a host of potentially threatening variables within the business case for the NBN.
If it is to attract private sector investment in the project, the government is going to have to explain how it can deliver high-speed broadband services, on a national basis, at prices that will be sufficiently competitive with legacy services to attract the overwhelming majority of broadband users.
This article by Business Spectator's Stephen Bartholomeusz is reproduced on ZDNet.com.au courtesy of a reciprocal publishing agreement.
7%
3%
I currently pay $60/month for ADSL2+ and $90.00ish per month for my fixed line phone service. (OK I'm either a Luddite or nuts: my wife and I each also have mobiles). Let's add $29 month for my mobile telephony (hers is in her business package) totaling $180 for IT charges per month.
ADSL2+ is broadband but only just. If an entrepreneurial reseller offers me a package of mixed connectivity like wireless + HS broadband (whose bandwidth will almost certainly dwarf that of wireless in perpetuity) both supporting VoIP, I will be willing to pay $150.00 2009 dollars for that service in 2015, while scrapping my copper wire connection.
What have I missed here.