Australia's planned National Broadband Network (News - Alert) (NBN) expects, over a decade, to build a fiber-based broadband access network providing 93 percent of Australian homes and small businesses with 100 Mbps service. Some locations will be served by wireless or satellite services that will operate at 12 Mbps. Fixed wireless will be used to supply service to about four percent of locations, while satellite is used to deliver service to about three percent of locations.
In total, about 13 million connections will be supplied.
"Retail pricing structure for fiber products is based around bundled (cheap or free) voice, fast broadband access and multi-channel TV," the NBN plan suggests. That little tidbit largely reflects the prevailing view that voice communications, though still a huge part of the overall value proposition, will not be the revenue driver for the network. Some might wonder about the relative contribution of multi-channel TV, over the medium or longer term, as well.
Of course, since the NBN will only supply wholesale access and transport, the specific retail plans will be determined by the retail providers themselves. Some may elect not to provide any one of the potential constituent services. Also, the NBN and its retail partners will continue to compete in a market with existing cable competition and expected growing competition from mobile networks as well.
One might guess, based on prior instances of robust wholesale regimes, that Telstra's current 70-percent-plus share of voice, and nearly-70-percent share of fixed broadband, could drop to about 40 percent, as already is the case in the wireless services domain. Market share of about 40 percent for fixed services would be consistent with other markets where robust wholesale competition is possible.
The big issue is how much the dominant fixed-line voice market might contract, though. Today, fixed voice represents about $13 billion in revenue, fixed broadband about $5 billion and wireless about $16 billion in revenue (all figures Australian dollars). The issue, of course, is that if the NBN itself expects voice services to be available "cheap or free," the $13 billion in voice revenue will shrink, with or without the new wholesale regime.
The retail and wholesale telecommunications market (fixed voice, fixed broadband, fixed data, wireless broadband, mobile voice and wholesale) is dominated by the incumbent, Telstra (News - Alert), with approximately 57 percent revenue market share. Optus is second with 25 percent.
About 10 percent of Australian homes already was "wireless only" in 2009, and the report suggests that could grow to 13 percent by the end of 2010. But the report also suggests that some households that are "wireless only" for voice still buy fixed broadband connections, suggesting that perhaps only about seven percent of Australian homes truly rely solely on wireless services.
That noted, the percentage of Australian customers buying wireless broadband has grown very rapidly since about 2008. The report indicates that 39 percent of Australian households now use a wireless broadband connection, while 61 percent use fixed broadband. One might suggest the possibility of significant wireless substitution, based on the already-high use of mobile broadband services.
But an argument can be made that mobile broadband will be complementary to fixed broadband, as mobile services will not be able to support the 100 Mbps speeds of the fixed network. The pricing of wireless broadband has successfully targeted the sub-$40, low end of the broadband market while offering comparable speeds to ADSL fixed broadband services, the report says.
"Much of mobile broadband demand is expected to be complementary to fixed line services, and over time it is anticipated that consumers’ demand for higher speed will translate into a greater dependence on FTTP infrastructure for new applications and services," the report assumes.
The plan assumes that wireless-only households (all wireless for all services) will grow to 16 percent by 2025 and stabilize at that level, remaining at 16 percent through 2040. Some will question that assumption. The plan also assumes "zero" wireless-only businesses and creation of new services that will drive buying of faster bandwidth products, though the plan does not quantify the revenue contributed by such future products, and cannot, at this point.
One might argue that there is room for consumers to substitute higher spending on broadband access, at higher speeds and higher prices, for less spending on voice services. The harder question is to assess whether the availability of 100 Mbps services will drive significantly-higher overall communications spending, and not just a shift from one product to another within a fixed communications budget.
"The main limiting factor in the early years of the NBN is expected to be the availability of applications that require high bandwidth," the report says. "Without these applications, consumers have limited reasons for migrating to the speeds offered by the NBN, and price becomes the main factor in driving consumer choices." NBN Co’s strategy is based on the expectation that as higher bandwidth becomes available, applications that take advantage of that bandwidth will be developed. But this "build it and they will come" assumption has proven wrong in the past.
It goes without saying that unless compelling new revenue-generating services are created, consumers will not have much incentive to upgrade their service plans much beyond where they already are, implying a tough slog to get average revenue per user up beyond $40 a month.
Some idea of the "retail pricing floor" can be gleaned from planned NBN pricing. Wholesale prices for a single 12 Mbps circuit are set at $24. A retail service provider will add operating, sales and capital costs to derive retail pricing. Other prices include wholesale charges of $27 for a 25 Mbps service with 10 Mbps return; $30 for a 25/20 service and $34 for a 50/20 access; $38 for 100/40 service.
Wholesale pricing for a 250/100 plan will cost $70; $100 a month will buy a 500/200 service and $150 is the wholesale price per month for a 1 Gbps/400 Mbps service. The charges intentionally are designed to encourage wholesale partners to buy and retail services running at 100 Mbps.
The NBN also will sell symmetrical services with guaranteed quality of service (committed information rates).
NBN Co will "provide a layer two bitstream service only, using a GPON (gigabit passive optical etwork) architecture. The company is not preparing for the provision of layer one services, layer one unbundling, functional or structural separation. Retail partners will not be able to buy "dark fiber," in other words.
Wholesale products will be sold supporting downstream bandwidths of 12 Mbps, 25 Mbps, 50 Mbps, 100 Mbps, 250 Mbps, 500 Mbps and 1 Gbps, with upstream bandwidths ranging from 1 Mbps up to 400 Mbps. The NBN also will offer wholesale voice capabilities.
The NBN will add video streaming delivery, but will not supply the rest of the video infrastructure. Also planned are features to support multi-location enterprises and 1 Gbps virtual LAN services, as well as protected diverse-routing services.
The plan calls for building 223,000 locations by June 2011, of which the bulk will be satellite connections. Fiber to home construction aims for 58,000 sites by June 2011, 259,000 by June 2012 and 952,000 units by June 2013, for a total of 1.26 million fiber-to-home connections by June 2013.
About 35,000 fiber connections are expected to be in service by June 2011, 102,000 locations by June 2012 and 374,000 locations by June 2013. Those targets are subject to both materials and labor constraints, though.
The NBN expects to have its own satellites in orbit and in operation by 2015, and will be leasing capacity until then.
The entire fiber network will take nine and a half years to build, assuming no materials or labor delays, and is projected to cost $36 billion. Revenues to 2020 are expected to be about $21 billion and operating costs are expected to run about $22 billion through 2020. The Australian government is contributing $27.5 billion, with debt financing of about $13 billion. The internal rate of return is expected to be seven percent.
For Telstra, the stakes are high, as Telstra will essentially be out of the infrastructure business, and purchase access and transport services from the NBN. Telstra also will divest its cable network customers as well.
Gary Kim (News - Alert) is a contributing editor for TMCnet. To read more of Gary’s articles, please visit his columnist page.
Edited by Tammy Wolf