What is the Role for Government in Broadband?

By Paul Fletcher - Australian Polity - Volume 2 (Number 2)

The Rudd and Gillard Labor Governments’ broadband policy is highly interventionist, extremely expensive and fundamentally damaging to competition — and the interests of broadband consumers.

Labor’s Interventionist Telecommunications Policy

Labor’s broadband policy puts government at the very centre. A wholly owned government company, NBN Co, is going to build and operate a fibre to the premises network serving 93 per cent of premises in Australia. In addition, this company will serve the remaining seven per cent of premises through building and operate a wireless network, and by purchasing and operating two satellites.

The policy is highly interventionist in several ways. To start with, it involves spending a huge amount of taxpayers’ money. Total spending will be well over $50 billion. Of this, $27.5 billion is a direct equity investment from government and a further $13.4 billion will be borrowed by NBN Co from the private markets (almost certainly requiring a government guarantee). In addition, NBN Co will pay Telstra some $9 billion under various contracts, again with such payments the subject of a government guarantee.

A further $2 billion is to be provided to Telstra directly by government (that is, by taxpayers). The largest part of this is to fund new arrangements for the universal service obligation, thus relieving Telstra of substantial costs. But there is money going to Telstra for a whole range of reasons — it even gets $100 million for retraining staff.

Secondly, the policy is extremely specific as to the kind of network to be built and the technology to be used. When then Prime Minister Rudd and Broadband Minister Conroy announced the new policy in April 2009, they committed that the network would be fibre to the premises to at least ninty per cent of premises, and use wireless and satellite in other areas. In other words, technology and deployment decisions were made not by technical experts but by elected politicians.

Thirdly, the policy directs the largest telecommunications company, Telstra, as to the areas in which it may, and may not, conduct business in the future. Telstra has in effect been forced to adopt a new business model: in future it will not provide retail services over its own network and will not continue operating a copper access network. Instead it will be required to purchase wholesale services from NBN Co.

This policy was implemented using a combination of legislated threats (unless Telstra co-operated it would be banned from bidding for wireless spectrum it needed to expand other parts of its business) and taxpayer funded inducements ($9 billion of payments by NBN Co and a further $2 billion directly by government.)

Fourthly, under the policy framework Labor has changed the law to make it effectively impossible for companies other than NBN Co to build and operate competing fibre networks. Labor has added new provisions to the Telecommunications Act which state that it is illegal to operate a network capable of delivering a ‘superfast broadband service’ (defined as 25 Mbps or more) unless you only deliver wholesale services over the network, and further that the wholesale services must be so called ‘layer 2 bitstream.’ In effect, this means that you cannot build and operate a competing network unless you adopt exactly the same business model as NBN Co.

Labors’ reversal in telecommunications

Labor’s NBN policy represents the reversal of a policy direction which has been followed in Australian telecommunications since the early nineties. That direction has been to liberalise the sector, to open it up to competition and to reduce direct government involvement.

Under this process, commenced by the Hawke and Keating Governments, the government owned monopolist Telstra was exposed to both fixed and mobile competition, initially in a period of managed competition beginning in 1992. During this phase, Optus entered the market as both a fixed and a mobile operator, and Vodafone entered as a mobile operator.

In 1997, the Howard Government built on these reforms by implementing the full liberalisation of telecommunications in Australia. Now, any interested party was free to build and operate either a fixed or a mobile telecommunications network.

The Howard Government also privatised Telstra — consistent with the reform path in many other countries. Typically, moving the dominant incumbent telecommunications company out of government ownership has been an important element of opening up the sector to competition.

The reform process clarified and distinguished the separate roles and responsibilities of telecommunications companies, and of government. Certainly, government has important roles to play, including the role of regulator. This includes technical regulation — such as ensuring that the network equipment used by different operators is compatible. It also includes economic and competition regulation — such as restraining dominant operators from, for example, charging excessive prices to smaller companies wishing to terminate calls to the dominant operator’s network.

Another role for Government is in funding telecommunications networks in areas where the private market is unlikely to invest. This typically means rural and remote areas — something of particular importance in a large and thinly populated country like Australia.

Conversely, what private sector telecommunications operators typically do better than government is market their products, innovate and respond to customer demand. A good example is the extraordinary growth of the mobile phone market in Australia. Mobile phone services started in the eighties; today mobile penetration exceeds one hundred per cent (that is, there is more than one service for every Australian). Over the last twenty five years, the rate of technical and service innovation, and network expansion, has been extraordinary — thanks in large part to vigorous competition between the three (and at times four) separate mobile network operators.

Consequences of Labor’s Interventionism

Labor’s NBN policy essentially discards the careful delineation of the appropriate roles of government and the private sector in telecommunications which previously applied. This has several undesirable consequences.

Firstly, by ensuring that all fixed line consumer telecommunications services are once again delivered by a government owned monopoly, Labor’s approach will result in a sector which is less efficient, less innovative, less flexible, less responsive to customer needs, and at a higher cost than a sector in which customers are served by privately owned companies.

Labor has deliberately put a monopoly at the centre of the new industry structure it is designing. NBN will own and operate the only significant fixed line network in Australia. Privately owned companies will deliver retail services using the NBN — but they will be at the mercy of the service quality, the responsiveness, and the pricing, of NBN Co. If NBN Co says it is going to take a week to connect a new service, then whether the retailer is Telstra, or Optus, or Primus or iiNet — the customer will be waiting a week for a new service.

While fixed line telecommunications in Australia today is far from being a perfect competitive market, Labor’s policy decisively weakens the competition that exists today. For example, around twenty per cent of Australian homes today can be served by the Optus cable network — giving them a choice of using that network or Telstra’s network for voice telephony and broadband.

Another important form of competition today comes from the use of ‘unbundled’ lines. These allow a competitor to Telstra — like Optus, iiNet or Primus — to lease the copper line running from a Telstra exchange to the customer’s home; install its own electronics in the exchange (together with a fibre optic connection from the exchange back into the competitor’s network), and serve the customer with much reduced involvement from Telstra.

Already nearly one million services are provided on this basis. Unbundled-based operators were the first to introduce higher speed ADSL 2+ (when Telstra was still only offering first generation ADSL.) Similarly, these operators introduced lower priced services such as ‘naked DSL’ (under which the end user can take a broadband service without having to pay another $30 or so per month for a separate voice service.)

Labor’s NBN policy will sweep away both of these sources of competition. NBN Co is paying Optus around $800 million to cease serving residential broadband customers on its cable network. It is also paying Telstra around $5 billion to shut down its network — meaning unbundling based services will be withdrawn from the market.

Today, Telstra has over 2.4 million broadband services and Optus has over 400,000 — but all of these customers will be forced onto the NBN and the two companies will cease providing broadband services over their networks.

Nor will there be much prospect of other players coming into the market to build new networks to compete with the NBN — thanks to Labor’s new laws making it illegal for any company to build and operate a ‘superfast carriage service’ unless it is wholesale only and replicates NBN Co’s business model.

The likely result is that the innovation and price competition we have seen in the broadband market — for example, from the unbundling based operators — will become a thing of the past.

A second dreadful consequence of Labor’s approach to the NBN is the hugely wasteful misallocation of public money. The cost has been steadily mounting since Labor first announced its NBN policy in March 2007. It then involved public spending of $4.7 billion and a joint venture with a private sector operator that would also contribute capital.

Next, in April 2009, Labor announced a much more expensive network design (fibre to the premises rather than fibre to the node) which was estimated to involve total spending of $43 billion — although we were told that up to half would come from the private sector, implying government spending would be around $22 billion. In 2010 we learned that the full cost of the NBN would need to be paid by the taxpayer — based on advice from management consultants McKinsey and KPMG that no private sector investor would be likely to put money into this venture.

We were also told, in mid 2010, of the $11 billion to be paid to Telstra — $9 billion from NBN Co and $2 billion directly from the government, but all of it ultimately coming from taxpayers. In December 2010, we learned that NBN Co’s Corporate Plan involved funding from government totalling $40.9 billion. In June 2011 NBN revealed plans to spend an extra $800 million with Optus. The taxpayer funded spending keeps gushing on.

A Better Approach

If we all agree that Australia’s broadband infrastructure needs to be upgraded, but if Labor is going about this task the wrong way — spending far too much money and sacrificing competition in broadband to the detriment of consumers — then what would be a better approach for government to take?

In my view, there are four key principles that government should follow: be much more cost effective in the use of taxpayers’ money, by targeting public funding for broadband to areas of market failure, especially in rural and remote Australia; rely on the private sector to build out improved broadband infrastructure where it is commercially attractive to do so (metropolitan Australia, where the substantial majority of Australian households are located); focus on getting the market structure right to stimulate competition; and set appropriate bandwidth targets (not technology targets) and establish mechanisms to ensure that private sector players meet those targets.

First, rather than committing to a massively expensive national infrastructure build wholly funded by taxpayers, government should prioritise funding in areas where the private sector cannot be expected to build. In the low population density areas of rural and remote Australia, it is difficult to make a commercial case for investment in new broadband networks. However, there are strong arguments for ensuring the availability of good broadband networks in these areas — including equity between city and country dwellers, and also the concern that if country areas do not have good broadband facilities this will hamper their ongoing economic development.

There is general agreement satellite and wireless networks are the technologies best suited to deliver broadband in remote and rural areas of Australia. The Howard Government was pursuing a wireless policy, committing funding in 2007 for the OPEL network. This was to deliver a nationwide rural wireless broadband network, initially delivering six Mbps and later rising to twelve Mbps. It was to be built and operated by the private sector but with a subsidy towards its capital cost of almost one billion dollars.

After the Rudd-Gillard government’s politically motivated 2008 decision to cancel OPEL, it later announced the rollout of a wireless broadband network for rural and remote Australia, using the same technology and radiofrequency spectrum which OPEL was to use. Rural Australians can be justifiably angry that Labor has delayed their access to wireless broadband by several years.

The next important principle is to maximise the role of the private sector. Building and operating broadband networks in densely populated areas such as the cities and metropolitan areas of Australia is a commercially attractive opportunity. For example, Telstra proposed to upgrade its network to fibre to the node to serve four million premises with a guaranteed twelve Mbps broadband service in the five biggest cities of Australia in 2005, entirely at its own cost. While the proposal did not proceed due to an impasse with government over regulatory issues, it is evidence of the private sector’s willingness to build and invest. In a world where the demands on the public purse are never ending, and a government’s spending capacity is finite, it makes very good sense to leave the construction and operation of broadband networks to the private sector whenever they are willing to build, thus conserving scarce public funds for other priorities.

Thirdly, government has a critically important role in maximising competition in the telecommunications sector and in ensuring the right market structure to achieve this outcome. The experience of the last fourteen years — since telecommunications was deregulated in 1997 — has taught us that the current market structure in fixed line telecommunications leaves too much power in the hands of the dominant incumbent Telstra.

The right policy response would be to structurally separate Telstra: to split it into two separately owned companies, one operating Telstra’s existing local access network (“Netco”) on a wholesale basis and the other operating the rest of Telstra’s businesses including its retail business. Netco would sell its services to all retail telecommunications companies including Telstra retail, and it would be prevented from selling retail services. With this structure in place, the unduly interventionist rules imposed by Labor could be swept away, such as the prohibition on other operators building broadband networks.

The fourth important task for government is to set national bandwidth goals, and to establish mechanisms under which the private sector operators in the cities (as well as the publicly subsisided operators in rural and remote Australia) will achieve these goals. Bandwidth goals should be specified in terms of the speed to be available to designated percentages of premises by designated dates. What government should not do is mandate particular technologies — this should be a decision for the organisations charged with meeting the public policy goals set by national leaders, drawing on their technological expertise and operational experience.

As to the right mechanisms to ensure the achievement of the bandwidth goals, these are likely to involve long term contractual agreements with network operators, backed by legislation. While the specific approaches to be used will depend on a range of factors including the attitudes of Telstra and other private sector operators, it is clear that one very important mechanism is setting the long term wholesale price. If Telstra is split into Netco and a retail business, then a key policy lever is the wholesale price which Netco is permitted to charge. This needs to be set at a level which allows Netco to achieve a return on its investment in upgrading its network to achieve the bandwidth goals. However it must also be at a level which ensures affordable broadband. A key factor here is commercial certainty — and hence setting a price for the long term.

Conclusion

Upgrading Australia’s broadband infrastructure is an important public policy goal. There is a central role for government in broadband policy. But it is not the role which Labor envisages: to build at huge expense a brand new super high speed network, to operate that network as an arm of government, and to use legislative force and public money to drive all existing and potential competitors out of the market.

Labor’s approach is not only appallingly wasteful. It is also very bad competition policy which entrenches NBN Co in a market dominant position and inevitably will expose consumers to higher prices, poorer service and weaker innovation than could be expected in a truly competitive market.

Instead, the proper role for government is to fund network build in rural and remote areas where there is no commercial business case; encourage the private sector to build in metropolitan areas where there is a business case; ensure there is strong competition, including through structurally separating Telstra; and set national bandwidth goals while establishing the right mechanisms to ensure that private sector operators build out their network to meet those goals.