Will NBN pricing go up and up?

Malcolm Turnbull’s latest piece of anti-NBN information came in the form of a Liberal party press release on 22nd August 2011, entitled Consumers will pay for Conroy’s roadmap to monopoly.

In it, he made numerous claims about future NBN pricing, referring to a discussion paper released by NBN co in July 2011, asking for submissions before it is submitted to the ACCC for approval.

So right off the bat, Turnbull’s implication that anything in this document is actually set in stone, is incorrect. It hasn’t even been submitted to the ACCC yet, let alone approved by them.

Pulling the PR apart, it doesn’t take long for the falsehoods to leap out. The first and second paragraphs are demonstrably false. He says:

“NBN Co is seeking the right to raise prices for most of its services – including broadband plans for business, streaming video, and any consumer plan offering faster downloads than today’s ADSL2 – by up to 5 per cent per annum more than inflation for the next 30 years……If NBN Co’s proposal to the ACCC is accepted, only its basic broadband offer (12 megabits per second downloads and 1 megabit per second uploads) will be subject to strict price caps until the late 2020s.”

The paper actually specifies the following, in relation to NBN’s wholesale pricing:

1. For the first 5 years, there can be no increase in price whatsoever for the AVC* entry speed (currently 12/1Mbps) or the most popular speed (whatever that may be). There can also be no increase in the CVC* component. (See below for an explanation of AVC and CVC)

2. For the next 25 years, the price of the entry speed (currently 12/1), the most popular speed (whatever that may be), and the CVC charge can be increased by no more than half of the CPI increase in any year. Cumulative increases are not permitted. NBN Co can apply to vary the contents of the “Price controlled product basket”, but any change must first be approved by the ACCC. Products outside the “price controlled basket” may be increased at a maximum of CPI+5%, but not if it would lead to the NBN making a profit.

3. The NBN is not permitted to make a profit. It must limit its revenue so that it does no more than break even.

In practise, these rules mean that the cost of the NBN for the vast majority of users is not permitted to increase at all for the first 5 years, and only permitted to increase at half the inflation rate for the following 25 years. This means that in real terms, the monthly price of the NBN service will fall each year by at least 50% of the CPI for the majority of users for at least the next 30 years.

Now let’s look at a few more of Turnbull’s statements:

“Between 1997-98 and 2008-09 inflation-adjusted prices fell 34 per cent for fixed-line telephone services and 49 per cent for mobile services, according to the ACCC.”

This is nothing more than a cherry-picking exercise from ACCC data. While some call charges have fallen, line rental has skyrocketed. Because the vast majority of Australians are required to pay line rental in order to get broadband (whether they want a home phone or not), this means the real cost of broadband has increased as well, not withstanding the increased data allowances and speed improvements that have occurred over time.

In 1998, telephone line rental was $11.65 per month. In 2011 it is $34.00 per month. That is an increase of just under 300% over a period when the average inflation rate was about 3% per annum. This year alone, Telstra increased line rental by 13.3%, thereby increasing the cost of phone and broadband for almost every Australian.

Turnbull writes:

“We know that lack of affordability is a critical contributor to the digital divide. We also know that eliminating competing infrastructure to protect a vastly over-capitalized statutory monopoly, as Labor has with the NBN, creates both the means and the motive for higher prices than a competitive market would deliver.”

“The whole justification given by the Government for building the NBN is to provide most Australians with access to super fast broadband at 100 mbps. And yet it is now quite clear that the consequence of spending over $50 billion will be that speeds comparable to ADSL2 will be available at similar or somewhat higher prices, but that the super fast speeds will only be available at higher and, over time, even higher prices making access to broadband less affordable.”

The NBN has neither the means nor the motive to increase pricing. We have already seen that they are proposing heavy restrictions on themselves for price increases, limiting the most popular products to increases of half the inflation rate, at most. Far from having a motive to further increase them, they will be required to set pricing so low that they do nothing more than break even. Contrast this to the current private sector, which expects to deliver profits of at least 15% on any project, and it’s not hard to see what will result in the lower pricing. Thanks to Telstra we can already see just how bad it would be, with their Brisbane fibre project set to charge prices about double the NBN’s while providing lower speeds. So much for the “lower prices of a competitive market”.

Of course, we know already that the $50 billion figure quoted by Turnbull is creative accounting at its best. However, the claim that the NBN will offer ADSL2-speeds at “similar or somewhat higher prices” is also demonstrably false. Thanks to Exetel’s NBN pricing release last month, we can now see that “ADSL-comperable” NBN speeds (which are actually about 50% higher than the average for ADSL2+) are available for some 30% less per month than current phone/ADSL2+ services. The news is even better if you’re currently a Telstra ADSL2+ customer. In your case, Exetel will give you 5x the speed and 40x the data for the same monthly price.

The NBN’s discussion paper is constantly at pains to point out that their primary goal is to set prices as low as possible in order to achieve the highest possible takeup rate. Far from Malcolm’s claim that they have a motive to increase pricing, the opposite is true. The lower their pricing, the higher the takeup, and the easier it will be for them to achieve their goals.

Further Information:

IT Wire: Turnbull’s NBN pricing scare gathers momentum

*AVC and CVC:

The NBN wholesale pricing charged to ISPs is made up of two major components, the AVC and the CVC. Each speed tier has a different AVC cost (eg: $24 for 12Mbps, $34 for 50Mbps). The CVC is a flat rate charged to internet providers based on the total speed they purchase to be shared by all of their customers. NBN Co estimate that the CVC component of their pricing typically accounts for between 1 and 3 dollars of the ISP’s cost to provide you with a service. This CVC component is part of the “price-controlled basket” irrespective of what user speed is chosen.

Will the NBN be too expensive?

(Updated 24th July. Check bottom of post for Exetel’s stunning NBN prices)

With the release today of the first real look at retail NBN pricing, there has been a flurry of activity variously saying how cheap the NBN is, or that it’s ridiculously expensive. So what are the facts?

I’ll cut straight to the chase. It’s a bit of both, but good news for most!

The first mistake most critics are making is to forget that Internode’s NBN plans include a phone line, and also $10 worth of phone calls. That means that you need to compare the NBN pricing to your your current broadband price PLUS phone line rental.

So let’s do some comparisons. First, let’s compare Internode’s own ADSL2+ bundles to their bronze NBN bundles:

Internode ADSL2+ Bundles

Internode NBN Bundles

As you can see, Internode’s entry level (12Mbps) NBN bundles are the same price as their ADSL2+ bundles. Even better, with typical ADSL2+ speeds in Australia being only 8-9Mbps, the cheapest NBN plan is (on average) about 50% faster than the ADSL2+ option, with the same data volume, for the same price.

There have been some concerns that the very bottom end of the market where price is most important, will be the losers under the NBN. So let’s compare some of the cheapest ADSL+phone bundles from an assortment of ISPs with Internode’s cheapest NBN+phone option. In each case below, I have included the cheapest available broadband plan, along with a phone line:

Cheapest NBN and ADSL/Cable plans compared

* ADSL2+ Speeds: Theoretically, ADSL2+ can do 24Mbps. However, in practice this speed is unattainable. Telstra quote a maximum of 20Mbps, and TPG only say “70% of their customers can access speeds of 10Mbps or more”. On average, Australians get a speed of 8-9Mbps over ADSL2+.

* Cable speeds: Theoretically, the standard Optus HFC cable speed is up to 30Mbps. However, in practice this speed is unattainable. Optus say “76% of cable customers can obtain speeds over 8Mbps”.

The above table quite clearly illustrates just how competitive Internode’s cheapest NBN plans are. The only non-NBN plan that appears to be better value is the Optus cable bundle, however this is only available to around 10% of the population.

Many of the on-net ADSL2+ plans are comparable to the NBN plan, sometimes with more data for similar prices, but without the included phone calls or the (future) nationwide coverage of the NBN plan.

However, the huge win for the NBN comes once you’re outside the 400-exchange coverage of the competitive ADSL2 networks. For the people that live in the ‘off-net’ areas serviced by the 4,600 exchanges where Telstra owns all the infrastructure, the NBN comes into it’s own and is undoubtedly better value than any current phone plus broadband bundles.

The competitive pricing is even more surprising considering Internode’s position as a ‘quality’ (as opposed to ‘budget’) ISP. This post will be updated as more ISPs release their NBN pricing.

Update 23/7/2011:

Budget internet service provider DoDo have announced today that they will be offering a “sub-$40″ NBN phone and broadband bundle as their entry level plan. Further details will be provided once they begin their NBN trial services.

Update 24/7/2011:

The news on NBN pricing just keeps getting better, with the release today of stunning prices from ISP Exetel, starting at just $34.50 per month for a 20GB, 12Mbps broadband service, with an optional VoIP telephone line for no additional cost. This is less than half the price Telstra currently charge for their basic phone/ADSL service today.

To put this further into perspective, NBN Co predicted a 12/1 plan with 50GB of data would cost $53-58 per month. Exetel offer this plan for just $39.50 a month.

Check out the full range of speeds and data allowances on the Exetel NBN website.

Can we leave the NBN to the private sector?

Opposition spokesman for communications, Malcolm Turnbull, often repeats the position that the Government should stay out of communications infrastructure, and projects such as the NBN should be left to the private sector (or “the market”), with some assistance provided for “unviable” areas.

So would such a proposal work? Has it worked in the past? Does it work elsewhere? Could we be better off?

Would a private company build the NBN?

Clearly, there is no private company (or consortium) that would build the Australian NBN as proposed by the Government. ie: A network covering 100% of the population, through fibre, wireless and satellite connections. There simply isn’t enough money in it for them.

For a private company to make an investment, they require a commercial return on that investment, because their primary objective is to earn a profit for their shareholders. Typically, the private sector demand a return of at least 12% per annum although often they won’t invest for a return of less than 15% per annum.

The NBN though, is projected to return just 7% per annum. About half the return demanded by the private sector. Such a level is fine for a Government because their primary objective isn’t to make a profit, but to deliver infrastructure and services. Generally Government projects don’t provide a return at all, so the NBN is well ahead of typical Government spending. The return means that it’s not really a spend at all, but an investment that will actually generate revenue and result in a valuable and saleable asset in the future.

Has the private sector succeeded up until now?

No. Or at least, not in Australia. We have amongst the slowest broadband speeds in the developed World. On the OECD broadband rankings for 2010, Australia ranked 18th out of 31 developed nations. For fibre connections, we ranked an even-lower 24th out of 31. Internet hosting giant Akamai reported that Australia ranked 51st (out of all countries) for Q4 2010 in average broadband connection speeds. This ranks Australia not only far behind the developed world, but also much of the undeveloped World! The ranking represented a fall of 11 places over the preceding 6 months as other countries improved their networks.

The primary internet technology in Australia is ADSL or its newer cousin ADSL2+. Even by 2011, Telstra have only upgraded ~1900 of their ~6000 telephone exchanges to ADSL2+. The situation for “competition” from other ADSL2+ providers is even worse, with only ~400 exchanges having ADSL2+ available from anyone other than Telstra. Newer, faster DSL technologies such as HDSL or VDSL have (with a few minor exceptions) not been deployed in Australia at all.

The rollout of HFC (cable) networks in Australia was halted over ten years ago, with only about 2.5 million premises passed, and a large portion of those unable to be connected due to technical limitations.

There has been no deployment of Fibre To The Premises (FTTP) networks in Australia except in a few new housing estates.

To summartise, while the rest of the World has made huge advances in the depolyment of ADSL2+, VDSL, Cable and Fibre, the “market” in Australia hasn’t even enabled 7-year-old ADSL2 technology in 2/3 of Australian exchanges! They stopped rolling out Cable internet a decade ago. There has been no attempt to upgrade existing areas to fibre optics. The market has quite clearly shown that it won’t even roll out cheap copper network upgrades, let alone consider deploying a world-class fibre network.

What about overseas?

The responsibility for building and funding telecommunications infrastructure varies around the World between Government, Semi-Government and the Private sector. For the sake of this point, I’ll concentrate on the USA, because their networks are almost exclusively built by the private sector and their geography and standards of living are similar to Australia. While broadband speeds in the USA aren’t exactly spectacular compared to countries such as Japan or South Korea, they are certainly well ahead of Australia. So if the private sector can build broadband networks there, why can’t they do it here?

The USA has enormous HFC (cable) networks covering over 80 million premises between them, plus a Fibre To The Premises network which passes about 15 million premises. Put together, this means about 80% of their population has access to fast broadband. All of these networks were built by the private sector without significant Government assistance.

There is one massive difference between the USA and Australia, when it comes to the viability of rolling out these broadband networks. CableTelevision. Because the networks don’t just deliver fast broadband, they deliver TV. In the USA, cableTV subscribers number almost 70 million (about 61%), while in Australia they are only about 1.6 million (about 16%) of premises (Both these figures exclude payTV delivered via satellite). So the telecommunication companies get two bites at revenue in the US, television and broadband. In Australia for the most part, they are limited to broadband.

With the strict anti-siphoning laws in Australia, cableTV will never achieve the subscription rates of the USA, and therefore there is no financial incentive for the private sector to roll out cable or fibre networks in Australia.

Would we be better off if the NBN were built by the private sector?

For the sake of argument, lets say the private sector did build the NBN….As mentioned above, they would expect a rate of return in the vicinity of 15%, about double the target set for the NBN. This could only be achieved in one of three ways:

1) Make the network smaller:

Networks get more expensive to build as the population density decreases. So an “NBN” built by the private sector would stop as soon as it was no longer able to achieve a 15% return. Although no company has publicly released an estimate of what this would equate to, we can look at the HFC network rollouts for a guide. These only pass about 25% of premises (Telstra) and 20% of premises (Optus). So we can safely assume that a private build would cover no more than 30% of the population, and probably considerably less than that.

Obviously, this would not make Australians better off!

2) Depend on Government subsidy:

Another option would be for the private sector to fund the profitable 30%, and the Government to subsidise them to build the remaining 70%. But this is Government expenditure, not investment. There would be no return on this spending, it’s simply money being given to the private sector to build the network. So while the private builder produces a healthy 15% profit, the taxpayer injects billions of dollars to subsidise the non-profitable construction. In this case, the private sector is certainly a winner. And Australians do get the ubiquitous fast broadband network, but it comes at a massive cost to taxpayers.

Obviously, this would not make Australians better off!

3) Charge higher prices:

If the private sector were to build the NBN as proposed, then the only way they could achieve their required return would be to charge higher prices for services on the network. In other words, customers would have to pay more.

Obviously, this would not make Australians better off!

Will the NBN be obsolete before it’s finished?

With the constant march of technology, there are often concerns expressed that the NBN may be obsolete before it’s even finished.

So what are the chances of this happening, and what technologies could potentially make the NBN obsolete?

Let’s begin with a basic explanation of the fibre portion of the NBN. It uses fibre-optic cable to transmit data in the form of light pulses. Fibre optic cables have been in service for about 40 years, and are the technology used to carry all the phone and internet traffic across and between countries. Our telephone exchanges are already linked by fibre cable, as are our wireless towers. Currently, there are only 6 major fibre optic cables joining Australia to the rest of the World, comprising a total of just 24 strands of fibre which carry basically all the telephone, FAX and internet traffic in and out of our country. What the NBN is doing is taking that fibre cable all the way to each house.

The strands of glass fibre used in NBN cables is technically the same as the fibre used in the cables joining countries together. While it won’t be used to carry the same massive data quantities, it is quite capable of carrying that much data in the future if required. The capacity of fibre cables is almost unlimited, and speeds are constantly increasing through the use of technologies such as Wavelength-division multiplexing.

The main system being used for the home NBN connections is called a GPON (or Gigabit Passive Optical Network). This allows for speeds of up to 1 Gigabit per second (Gbps) and is currently the fastest such technology available. If faster versions of GPON are developed in the future, the NBN can be upgraded in a similar way as the copper network was upgraded from ADSL to ADSL2+. In other words, the cables remain the same, but the equipment is upgraded to take advantage of the faster speeds.

However, bear in mind that 1Gbps is the fastest internet system being implemented anywhere in the World, and is 100 times faster than the current average broadband speed in Australia. Even if faster GPON systems become available, the NBN will certainly not become obsolete.

Contrary to erroneous claims, fibre optic cables have no specific lifetime or known failure mechanism. There have been no widespread failures of exisCorning, one of the World’s largest manufacturers (and the supplier of NBN fibre cables), report that they have fibre cables submerged and under tension, and still working flawlessly after 35 years of constant testing. They also report that cable construction has improved over that period, so cables manufactured today are even more durable. Other manufacturers predict an average life of 60 years.

Could copper provide faster speeds than the NBN?

No. There is no copper technology in use or under development that can provide the speeds available over fibre. This is why telecommunication companies replaced their copper inter-exchange links with fibre decades ago. Fast copper technologies (such as VDSL2) are able to provide good speeds over very short distances, but they require two phone lines in excellent condition to do so. There is no prospect that any copper technology could overtake fibre for speed, primarily due to physical limitations of copper wire, including resistance, attenuation and signal-noise ratio.

The other copper-based technology is Hybrid Fibre Coaxial (HFC) cable. This is the pay TV cable used in some areas of Australia. It uses fibre for the main ‘trunk’ connections, and copper co-axial cable for the ‘branches’ down side streets. In some areas, it is capable of delivering 100Mbps to be shared amongst users on each branch. There is no prospect of HFC ever delivering speeds that exceed the NBN’s 1Gbps.

Could wireless broadband provide faster speeds than the NBN?

No. Unfortunately, the speed and capacity of wireless networks is hampered by unassailable limitations of the technology. It is a physical impossibility for any wireless network to be capable of exceeding the speed or capacity of a fibre-optic network.

Each time we increase the speed of a wireless network or add more users, it consumes more of the radio spectrum. There is only a very small amount of radio spectrum available for use in wireless broadband networks, because other frequencies are reserved for use in broadcasting, CB radios, WiFi networks and assorted other communication systems. The available spectrum for wireless broadband must then be shared amongst all the network operators. For example, if there is 300MHz available, Telstra may own 100MHz, Optus another 100MHz and Vodafone the final 100MHz. Delivering high speeds then becomes even more difficult, because it is rare for any one operator to own continuous blocks of frequencies, which is what’s required to deliver higher speeds.

However, even if you could eliminate all other radio uses and allocate everything to one wireless operator, a single strand of fibre optic cable can still carry thousands of times more information that the entire radio spectrum put together.

Another fact to keep in mind is that when you hear the speeds of wireless networks being reported, the quoted speeds represent the theoretical capacity of each cell of the network. These speeds are never achievable in practice, both due to technical limiting factors (such path loss from obstructions, distance and weather), and by the number of users connected to the cell at any one time. Typically, users achieve maximum speeds of about 1/10th of the quoted network speed, but often far less than that. The more users on the wireless network, the slower it operates.

As previously discussed, there is not a single country or telecommunications company anywhere in the World that is proposing a wireless network to replace their fixed network in urban areas, including the United States.

There is much more information on wireless in the following articles:

• Broadband Facts, Fiction and Urban Myths

• NBN: Why Not Wireless?

There is no technology currently available, in development, or even on the drawing boards that can make fibre-optic cables obsolete.

Angry ISPs?

There have been some concerns expressed recently that the NBN will drastically reduce the number of Internet Service Providers (ISPs) in Australia, because the system will make it too expensive for smaller operators to compete with the larger ones.

Simon Hackett, the managing director of Internode (An NBN proponent), suggested that it would not be financially viable for any ISP with fewer than around 250,000 customers to offer NBN services nationally without the assistance of a larger provider acting as a wholesaler.

Seemingly on-cue, Optus announced that they would offer wholesale services on the NBN to smaller providers.

Before explaining the issues involved, let me change some terminology. Under the NBN, ISPs will instead be known as RSPs (Retail Service Providers), because they will offer telephone and other services as well as internet services. So from here on, I’ll use the term RSPs to describe these companies.

Background

The issue for the smaller RSPs relates to the number of Points Of Interconnect (POI) for the NBN. A POI is a place where the NBN leaves off, and the RSP takes over. The NBN will have a number of POI across the country, and for an RSP to offer services Australia-wide, they would need to connect their own “backhaul network” to every POI. There are two points of view with regard to the number of POI that should be implemented across Australia:

Small number of POI:

The NBN Co had originally proposed just 14 POI across the country. This position had the support of most of the smaller RSPs, but drew strong criticism from larger RSPs.

Pros:

• Lowers the cost of offering a nationwide service, leading to a greater number of RSPs.

• Simplifies the network design for NBN Co.

Cons:

• Leads to the overbuilding of existing infrastructure, rendering it essentially worthless.

• Potentially reduces competition, because there is no prospect of cost savings associated with RSPs having their own, efficient backhaul network.

Large number of POI:

Larger RSPs, most notably Telstra and Optus, strongly lobbied for a large number of POI to maximise the return on their existing infrastructure. Calls from these larger companies varied from 200 up to 500 POI to be placed around the country.

Pros:

• Allows the use of the substantial existing networks of larger RSPs such as Telstra, Optus and iiNet. Potentially reducing the costs of these operators, potentially reducing retail pricing.

Cons:

• Drastically increases the cost of providing a national service for those operators who do not have a large existing network.

• Potentially reduces retail competition due to reduced number of players.

In the interests of fairness, the Government decided that the Australian Competition and Consumer Commission (ACCC) should review the options and decide on the number of POI that the NBN should provide. After completing their inquiry, the ACCC took a “middle ground” position between the demands of the small and large RSPs, and directed NBN to build 120 POI (later increased to 121) across the country.

 

The upshot of the POI decision:

It should be clear from the above, that no matter what the ACCC decided in relation to POI, there would be winners and losers in the RSP industry. In the end, they went with a compromise solution that makes it viable for many RSPs to make direct connections to the NBN, while allowing smaller ISPs to continue to operate nationally by taking up wholesale services through the larger RSPs. Additionally, smaller RSPs who wish to remain in a certain area (eg: Adelaide’s Adam Internet) are able to connect to just the POI they wish to, with the potential to expand their customer base by gradually adding new areas or taking up wholesale services for those areas where they only have a small number of customers. The NBN Co’s Jim Hassell says that the NBN Co want to encourage such regional RSPs to thrive.

The POI issue is an inevitable part of any major telecommunications upgrade. No matter what technology is used or how it is implemented, there will always be winners and losers during such a massive shakeup of the industry.

Isn’t the USA doing a wireless NBN?

Barely an NBN debate goes by without someone saying “Why is Australia laying cables, when the USA have decided to go wireless?”

Demonstrably false comments by AM radio “shock jocks” and conservative newspaper columnists are propagating this incorrect belief.

Let me be absolutely clear: The USA is not proposing to build a wireless network to replace their fixed networks.

Because, as they well know, it’s a physical impossibility for them to do such a thing.

So, where did this fallacy originate?

In his State of the Union address, President Obama announced that the US Government would assist the development of 4G wireless broadband networks by freeing up additional radio spectrum. This was necessary in the USA for a couple of major reasons:

• Due to a lack of radio spectrum, the wireless broadband networks in the USA are in serious trouble, and it has been predicted that they will collapse by 2013 unless more spectrum can be found.
• The largest 3G wireless broadband network in the USA only covers 75% of the country’s population. This it pitiful compared to Australia, where Telstra cover an incredible 99% of the population with their 3.5G network. The lack of coverage in the USA makes it difficult for emergency services, for example, to access data wirelessly across the country.

But the wireless announcement was just a small part of the wider US broadband plan to drastically improve the country’s networks. The plan includes a target to provide 100 million US homes (about 70% of their population) with access to broadband providing actual download speeds of 100Mbps within 10 years. They have also set a goal for all schools, hospitals and public buildings to access speeds of 1Gbps, which can only be achieved using fibre-optic cables. The US already has a large 150Mbps fibre-optic network that currently provides coverage to 15.6 million homes. They also have massive HFC cable networks which pass over 80 million homes between them. These networks can be upgraded to provide the 100Mbps speeds relatively cheaply, but the 1Gbps fibre connections will have to be constructed.

So what the USA have is almost the opposite of Australia. They already have fast, fixed broadband networks which can be upgraded to provide speeds of around 100Mbps. While only 1/10th of the speed of the Australian NBN, it is within the scope of the President’s target. Their wireless networks, however, will require large public investments and incentives to bring them up to scratch.

Whether the journalists are being deliberately deceptive, or they just don’t bother reading the entire US plan is unknown. Either way, claims that the US are proposing or building a “wireless NBN” to replace their fixed networks is absolutely false.

Will the NBN be a new monopoly, just like Telstra?

One of the most common concerns about the NBN is that the Australian Government is replacing one monopoly -Telstra- with another, NBN Co.

While on the face of it this may seem reasonable, there are major differences between the organisations and how they will operate, which will prevent NBN Co ever resembling Telstra.

Telstra is a profit-driven, vertically-integrated monopoly. In simple terms, this means that they are both the owner/wholesaler of the network, plus a retailer of services on that network. This creates a very poor situation for consumers, because to protect their high retail profit margins, Telstra are able to restrict access to their network for other Internet Service Providers while also setting both the wholesale and retail pricing for services.

For example, in order to prevent being forced to allow wholesale access to ADSL2+ infrastructure, for several years Telstra openly refused to switch on the services, even though they had already upgraded their telephone exchanges and the technology was widely used overseas. This forced ISPs to continue to use the old ADSL1 network and pay the higher prices for accessing it. In the end, Telstra won agreement that they could have exclusive use of that part of their network, which means that only a fraction of telephone exchanges across the country have access to ADSL2+ from anyone but Telstra. This has effectively eliminated ADSL2+ retail competition at about 8700 of Telstra’s 9200 telephone exchanges. Telstra continue to request exemptions for access to their network from other providers, further reducing retail competition. A vertically-integrated monopoly is never good for consumers.

The difference with NBN Co is that they are a infrastructure-only monopoly, with ACCC-monitored uniform pricing and wholesale-only open access, irrespective of whether a customer is on a fibre, wireless or satellite connection. Because NBN Co are not permitted to be a retailer (This is enshrined in legislation), they have no high-margin retail arm to protect. They are required to give access to any Retail Service Provider (eg: Telstra, Optus, iiNet, TPG etc) at the same price. They cannot give bulk discounts or special deals.

The fact is that telecommunications infrastructure will always be a monopoly, because it’s uneconomical to run out duplicate networks. Optus tried it in the 1990s, and it cost them hundreds of millions of dollars every year. If you allowed different companies to build in different areas, that would solve nothing on the monopoly front. Each area would simply be served by a different monopoly, with no competition with those areas.

Think of the NBN Co as a electricity or water company. It doesn’t make sense to have multiple power lines or water pipes running to every house any more than it makes sense to run multiple communication lines to each house. So you build one lot of infrastructure, regulate the pricing, and then allow heavy competition by retailers to access the network and provide different service solutions to consumers.