Broadband prices are the lowest they have ever been in Australia. But for how long?

The National Broadband Network:
Optus' viewpoint of what must happen to
boost broadband competition and
benefit all Australians May 2008

1. Introduction

In its March 2007 policy document "A Broadband Future for Australia - Building a national Broadband Network" the then Labor Opposition committed to roll out a national high-speed broadband network as a means of stimulating productivity, competitiveness and wealth creation. The Rudd Government has set the delivery of this network as one of its key policy goals of its first term in office.

However, delivering the network is only one part of the equation. Australia will capture very little benefit from this significant new piece of infrastructure unless it delivers innovative services at the lowest possible prices to the broadest possible base of consumers. This will only occur if there is vigorous and effective competition in the delivery of broadband services over this network.

There is a clear and present danger that in its rush to roll out the National Broadband Network (NBN), the Government will give away critical competition and consumer safeguards. This would see Telstra regain its monopoly control of fixed line voice and broadband. Such an outcome will be disastrous for Australia.

The key challenge for Government is to set the correct regulatory framework for access to the NBN. This will be fundamental to determining whether the NBN delivers the benefits which the Government anticipates.

2. Why Regulatory Reform is Critical

The Government has established two quite separate processes to give effect to its NBN policy:
(a) Selection of a successful bidder to construct the NBN; and
(b) Developing the regulatory framework that will facilitate competition on the NBN and deliver the desired consumer benefits.

The regulatory framework must be determined quite independently of the identity of the successful bidder. There are key principles which must apply whoever builds the network. If they are not agreed, no new network should be built.

Conversely, if Government sets the correct regulatory framework for the NBN such that it maximises competition and consumer welfare, then access seekers and consumers are likely to be indifferent as to who builds the network. This fact was acknowledged for example by the ACCC Chairman, Graeme Samuel, in an interview on 15 March with Communications Day:

"I don't think that actual ownership of the network .is all that relevant to what we will ultimately be dealing with which is the regulatory environment that will impact upon the broadband network as it's rolled out".

In developing the regulatory framework to apply to the NBN, Government needs to take account of:

• The importance of competition in driving broadband uptake.
• The economics and architecture of the proposed fibre to the node network;
• The relative vigour of competition under resale and unbundling; and
• The historic problems arising from Telstra's vertical integration.

3. The Importance of Competition

3.1 Emerging competition is delivering consumer benefits

The stand-out regulatory policy success of the last decade has been the requirement for Telstra to unbundle its local copper loop network. This has enabled competitors like Optus, Primus, Internode and iiNet to lease Telstra's last mile copper loop (the ULLS or LSS service) and combine it with their own electronics in the exchange. This lets them compete directly with Telstra.

Today's network with unbundling

At the start of 2008 some 1084 competitor DSLAMs were deployed across metropolitan Australia¹ in some 387 exchanges. They were used by Telstra's competitors to serve almost 800,000 customers. This in turn has driven important benefits to consumers - lower prices, better service and greater innovation. These benefits result directly from the existence of unbundling and the competition it has unleashed. Had Telstra's competitors relied solely on 'reselling' Telstra's wholesale DSL services, they would not have been able to deliver these important consumer benefits.

Competitors are using their own infrastructure to deliver innovative services such as Optus' Fusion product ($79/month for broadband plus telephony with unlimited standard local, national long distance and calls to Optus GSM Mobile) and iiNet's Naked DSL ($49.95 for broadband - without the requirement to pay for line rental).

This competition is helping Australia's broadband market to catch up with the world, recovering from a delayed and sluggish start. The chart below shows how growth jumped sharply once competitors such as Optus entered the DSL market.

The clear competitive benefits of unbundling have been recognised by the Chairman of the ACCC, Graeme Samuel, in a recent speech to the Australian Telecommunications Users Group:

"Increased competition in the provision of broadband services has seen progressively lower broadband prices, increased data caps, better speeds and new innovation and products (such as naked DSL). This increased competition in broadband by other ISPs and carriers owes a significant debt to being able to obtain access to Telstra's copper loop. Competitors have this access through the declaration of the unconditioned local loop service (ULLS) and the line sharing service (LSS)"².

3.2 NBN Threatens Competition

However, the government's objective for a ubiquitous broadband network based on fibre-to-the-node (FTTN) technology represents a real threat to the emergent competition in the local loop. FTTN is an ideal technology for an incumbent with anti-competitive ambitions.

Unlike the existing copper network, FTTN cannot be unbundled. If Telstra is able to build the FTTN network on the terms it proposes, it will be protected against competitive entry. Telstra will be the monopoly supplier - allowing it to keep prices much higher than in a competitive market.

Secondly, this means that the only competition will be from resellers - but with the speed, grade of service and other features all determined by Telstra, it will be impossible for resellers to differentiate or to offer truly effective competition.

(i) NBN Cannot be Unbundled

The focus of interconnection today is the Telstra exchange, where competitors can locate their electronic equipment to access the Telstra copper loop. But things are very different under FTTN architecture because the Telstra exchange is by-passed and the electronic equipment is located in a street-side node or cabinet closer to the customer.

The NBN design makes the existing exchange equipment - both Telstra and competitor owned - essentially useless. It also means that to 'unbundle' a FTTN network, competitors would need to put new equipment into each node. This is possible in theory; but for both commercial and technical reasons it is not viable in practice. In turn, this means there will be only one physical NBN. The following diagram helps to identify the problem.

Under Telstra's proposed FTTN - competitor infrastructure is stranded

(ii) Competition will revert to resale - which is much weaker

Since unbundling of the FTTN is not viable, the construction of a FTTN-based NBN will stop dead the current trend towards stronger competition based on unbundling mandated by Government. Instead the Australian broadband market will be forced back to a resale based model.

The prospects for competition in such an environment are poor unless there is very profound corrective action by Government. Australia has had considerable experience of resale competition in fixed line voice telephony on terms dictated by Telstra. This experience shows that competitors operate on very low margins, so they are unable to put the incumbent under much price pressure.

Equally as important, resellers are unable to differentiate their product from Telstra's. Under an FTTN network, if competitors are reselling Telstra's product, they will have to accept Telstra's decisions regarding bandwidth, grade of service, and other key factors such as the contention ratio. This is because all these decisions are made by configuring the node in one way or another - and Telstra will control the node.

More concerning is that an NBN based on the model above, owned and operated by Telstra, will enable Telstra to regain its monopoly control of the local loop. In turn it will be able to exploit that control to the detriment of competition.

(iii) History shows how Telstra will quickly exploit its monopoly position

The past decade provides far too many examples of Telstra using its position as the vertically integrated dominant supplier, to undermine competition in the provision of fixed line services.

This problem was particularly acute before unbundling started to become effective early in this decade. Before that time, competitors had to resell Telstra's services with limited capacity to differentiate or design their own product offerings. The following table provides three real and practical examples of the problems faced by access seekers competing against Telstra.

Table 1

Broadband emerged very slowly in Australia. This was mainly due to Telstra's ability to use its position to undermine competition. It was in Telstra's interest to slow-down the pace of broadband take-up since it threatened to cannibalise its existing lucrative revenue stream from dial-up internet services. Telstra reaped the benefits of the status quo which saw many customers renting a second line from Telstra for their internet service. The development of broadband threatened this cosy position since customers migrating to broadband were likely to cancel their second line.

Telstra specifically used its vertical integration to chill competition in broadband services and hence delay the take-up of broadband. The way it did this was to;

(a) Initially set high prices for broadband access. Telstra only dropped its retail prices for broadband services when Optus commenced re-selling its wholesale DSL service in February 2004;

(b) Squeeze the margin available to competitors taking its wholesale DSL service (see table above);

(c) Limit the functionality of its wholesale broadband services and thus make it harder for competitors to innovate (for example, by offering different versions of the service such as symmetrical DSL); and

(d) Artificially cap broadband speeds at 1.5Mbps on first generation ADSL (a technology capable of up to 8 Mbps) and delaying until February 2008 the introduction of ADSL2+ services (except in exchanges where it faced direct competition from ADSL2+ services provided by competitors using ULLS access to deploy their own equipment). It could have turned on ADSL2+ in hundreds of new exchanges by simply flicking a switch (as is demonstrated by the speed with which Telstra turned on exchanges following its February 2008 announcement) but for many months Telstra refused to take this simple step towards improving services to its customers.

The Government and the ACCC have struggled to prevent these attacks on competition and to control Telstra. The legal powers they have are simply too weak; and Telstra has enormous capacity to use legal action to resist any enforcement action by Government.

The ACCC recently noted that it is currently involved in 47 legal actions initiated by Telstra. This includes:

(a) 1 appeal to the Full Federal court;

(b) 12 ADJR actions in the Federal court;

(c) 1 Federal court ADJR action regarding administration of retail price controls; and

(d) 33 applications to the Administrative Appeals Tribunal for review of ACCC decisions on Freedom of Information requests (some see this as an abuse of the FOI provisions).

The central problem is that as a vertically integrated supplier with a dominant market position Telstra has strong incentives to engage in either anti-competitive conduct or to game the regulatory processes since actions will help to protect its position.

(iv) The impact to consumers and businesses could be disastrous

Without the correct regulatory settings for access to the NBN, competition in fixed line voice and broadband services will stop dead in its tracks. Unbundling-based competition will cease. There will be nothing to replace it on the NBN.

Telstra fully understands that NBN gives it a golden opportunity to kill competition stone dead and reinstate its monopoly. Telstra has publicly stated that it plans to sharply increase its profits, boosting its EBITDA margin to 50% from the current levels of around 42%.

Such an sharp jump in profit can only be achieved by charging customers more for the voice and data services they get today. This fact has been publicly acknowledged by Telstra on several occasions. Dr Phil Burgess, Telstra's Group Managing Director, Public Policy & Communications recently said that Telstra aims to be "a premium provider charging premium prices". He also indicated that Telstra required a return on any investment in NBN "north of 18%". Such a return is massively in excess of the returns normally allowed by the ACCC in setting prices for services delivered over regulated assets. If Telstra were able to charge such a rate, it would recoup its investment in under 5 years. Yet the network will last 10 to 15 years at least. This can only mean high prices and a long period of monopoly profits for Telstra.

The stark outcome for consumers under the model Telstra propose is clear;

• Competition will be stifled;
• Prices will rise significantly;
• Innovation will be muted;
• Take-up of high-speed broadband services will be held-back;
• The IT revolution in schools will be threatened; and
• Productivity gains will not be realised.

In contrast, Telstra will be enriched as competition is marginalised.

4. What Must Happen: Structural Separation

(i) Telstra's Game Plan Must Be Blocked

Telstra has attempted to capture the Government in a pincer-like grip to ensure that its interests prevail. On the one hand it has talked-up the need for an FTTN roll-out and offered the prospects that it will build out such a network rapidly - but only if it receives significant regulatory concessions (such as high access prices and a weaker regulatory regime with no role for the ACCC). On the other hand Telstra has sought to discourage Government from choosing another party to build the NBN by threatening to launch the "mother of all" legal blockades. In parallel, it has sought to demonise one of the truly independent voices in the debate, namely the ACCC.

Telstra must not be allowed to succeed. The Government can have an NBN and preserve competition on that network. However, this requires a radical paradigm shift to the way that telecommunications is regulated, otherwise the benefits of the NBN will not be realised and consumer interests will be harmed.

(ii) Structural Separation is Key to Preserve Competition

Simple economics makes it virtually inevitable that the NBN will be a monopoly infrastructure. The appropriate policy response to preserve competition in such an environment, regardless of who builds the network, is to ensure "structural separation". This means that the entity which owns the network must be different from the entities which sell services delivered over the network in the retail market. This is a necessary condition to achieve a competitive and efficient telecommunications market.

Under structural separation, the entity owning the NBN would have separate facilities, systems and staff and separate ownership from any retail operator, be that Telstra or any other operator. As a standalone business the incentives of the NBN owning entity would differ from those faced by Telstra today.

In today's world, Telstra is vertically integrated. It is the market leading retailer with a market share of around 80%. It also owns the 'bottleneck' asset, the 'last mile' copper network, or 'access network', which connects to every home and business in Australia. Any competitor to Telstra must use that network if it is to connect to customers - and in turn it must purchase from Telstra the right to use that network. But of course in selling such services, Telstra as a vertically integrated operator has the incentive - and the means - to discriminate in favour of its downstream businesses. This leads to both price and non-price discrimination. Some examples were given earlier in this paper.

Under structural separation, this problem is solved. The entity which owns the access network (which in the new world will be the NBN, that is, a high speed national broadband network delivering speeds of up to 40 Mbps) is different to the downstream retail operators. That means that it would have a strong incentive to sell to all downstream retail operators, so as to maximise its own profits. Unlike the vertically integrated Telstra of today, it would not have an incentive to discriminate in favour of one particular purchaser of wholesale services (namely, Telstra Retail) and against all others. Of course, the owner of the NBN would certainly require continued regulation: it would be the monopoly provider of fixed line high-speed broadband services and absent any regulation it would seek to charge unacceptably high prices (just like any other monopolist).

But the structurally separated network would not need to be subject to a whole layer of regulation that would be needed if Telstra were to build the high-speed broadband network as a fully integrated provider.

The policy case for structural separation of Telstra is a strong one. It has always been a strong one, but it is made greatly stronger by the arrival of the NBN. There is a compelling case that neither Telstra nor any other party should be permitted to build an NBN unless it is structurally separated. Structural separation would mean that the high-speed broadband network owner would:

• Have no incentive to engage in price or non-price sabotage against a particular access seeker;
• Engage in efficient pricing at the wholesale level rather than at the retail level. This ensures that all access seekers face the same true economic wholesale prices;
• Provide all access seekers with equal access to information important to their planning processes; and
• Have a lower cost of capital revealed in financial markets reflecting its lower risks as a standalone network owner.

Structural separation changes the incentives. In turn this removes the need for extensive and detailed rules to block the obvious anticompetitive strategies that the vertically integrated NBN owner would otherwise pursue. The market structure will naturally deliver pro-competitive outcomes. These benefits have been recognised by the ACCC:

"a vertically separated ownership model could reduce incentives for the access provider to discriminate between downstream users of the access service and, therefore, facilitate strong and effective competition between access seekers in retail markets."³

Telstra has argued strongly against structural separation. This is not surprising because separation would greatly weaken its ability to engage in anti-competitive activity. Today, resources and capital are allocated between Telstra's retail and network division by management fiat. In a structurally separate environment, Telstra's business divisions would be separate companies that would need to source capital from markets which would judge them on their true business performance. Their individual business units would not be 'protected' from competition by anti-competitive strategies and as such they would, perhaps for the first time, face competition on their merits.

There is increasing evidence to suggest that when the opportunities to engage in anticompetitive conduct are reduced, generally through tight regulation and effective operational separation, investors are recognising the gains from structural separation. Evidence is becoming clear in Ireland and New Zealand.

In summary, vertical integration of Telstra creates insidious incentives to engage in anticompetitive conduct because the vertically integrated Telstra will control access to an essential input into downstream markets which are potentially competitive. Such incentives cannot be readily overcome by regulation. Therefore, it is essential for Government to use the tender process to structurally separate Telstra.

5. Conclusion

Australia faces a clear choice for its broadband future.

The wrong choice is for the new network to be built and owned by a vertically integrated Telstra. Even with today's network architecture, Telstra has used its dominance to stifle competition. With the NBN, where there will be no unbundling to maintain competitive pressure, the consequences would be even worse than today. Prices will be high, service will be poor and take up of services on the NBN will be well below their potential.

The right way forward is for Australia to get a new national broadband network - along with structural separation. This will deliver the best of both worlds: a world leading high bandwidth broadband network; plus vigorous competition at the retail level to ensure the lowest possible prices, the best possible service, and the greatest possible degree of innovation.

Government must choose the right way forward, otherwise Australia will end up with broadband dominated by a vertically integrated monopoly. That would be a significant backward step compared to the increasingly vigorous competitive landscape of Australian broadband today.