Australia's FTTN project: something's got to give

David Kennedy, Research Director

The Australian government has launched a tender process to invest up to A$4.7 billion of public funds for the construction of a FTTN network to reach 98% of the population. The balance of the funding is to be provided by the winning tenderer. So far, it appears that Telstra, and the Terria consortium led by Optus will enter the race. Optus also individually lodged the required A$5 million bond last week, and we await confirmation on other bidders.

In recent months, indications from the government had been that the total cost would be around A$8.5 billion. But Telstra CEO Sol Trujillo stated that a A$15 billion price tag would be more realistic. In one of life's little ironies, the government originally costed its proposal on the basis of Telstra estimates made back in 2005.

The goalposts have now shifted, and this will require the government to adopt a flexible approach. There was always scepticism that 98% of the population could get FTTN technology at the proposed level of investment. An extensive programme of fibre rollout in regional and rural Australia would be needed to reach so many premises at the data speeds the government wants. The Australian population is highly urbanised, but there is a 'long tail' of towns, settlements and dispersed outer suburbs that will be expensive to provide with fixed broadband, and it certainly amounts to more than 2% of the population. The proposed level of investment and the ambitious 98% target don't seem to add up. Something has got to give.

The rethink about costs comes at an awkward time for both the government and tenderers. In the recent Federal Budget, the government rolled the FTTN funding into a larger fund that will also be available for road and rail investment. This will increase contestability, and put additional pressure on FTTN tenderers to minimise their call on the fund.

This adds to the challenges already facing bidders. The regulation of wholesale network pricing will reduce returns, and the likely imposition of operational separation requirements will raise costs. Of course, extending FTTN into marginal or even unprofitable areas will further erode the overall rate of return.

Adhering to the 98% target with FTTN technology would require the government to consider some unpalatable options. It could significantly increase the level of public funding, or modify its demand for a commercial return on its investment. The problem is that potential investors in road and rail infrastructure would justifiably cry foul (not to mention the Treasury). Alternatively, it could allow the winner to charge rural customers more than urban customers, but this would be politically unpopular.

These dilemmas are a consequence of the one-size-fits-all national approach the government initially adopted. It is important to realise that similar publicly funded fibre networks in other places have been much more targeted. The two exemplars, Amsterdam and Singapore, are small high-density markets where the business case for fibre is relatively easy to make. The diversity of the Australian market suggests that a different, more regionalised approach would work better. 'One technology to rule them all' is not the best approach for a continental network.

Recently, the government has shown signs of a more flexible approach. Although clarification of the tender calls for a network 'that uses FTTN or FTTP architecture', this leaves room for wireless access. The Terria consortium has hinted that wireless and satellite will form an important part of its bid.

In another one of life's little ironies, the new government recently cancelled a contract with an Optus-led joint venture to build such a network using WiMAX. However Ovum thinks that some kind of wireless solution is inevitable outside the areas where FTTN is commercially viable.

However, there is plenty of 3G mobile broadband capacity being installed on a commercial basis in Australia's hinterlands. This could be exploited to offer both fixed and mobile broadband services, if the required spectrum and backhaul capacity is made available. Using this existing commercial infrastructure would also reduce the costs of the project.

The government's tender process can accommodate this complexity, provided that achieving the 98% target with FTTN technology doesn't become an end in itself. A little pride may need to be swallowed, but the desired result is still achievable with a flexible approach.

David Kennedy is a research director responsible for broadband and wireline research. Based in Melbourne, Australia, he co-ordinates Ovum's broadband research priorities across Australia, New Zealand, South-East Asia and Greater China.