This week Malcolm Turnbull confirmed announced he had asked for the resignations of the NBN board. He stamped his authority on the direction of the network through a letter detailing government expectations of NBN Co, and revealed the reconstituted board would have 60 days to review the rollout and cost of the network. Unlike the last three years of election campaigning, this was a calm, pragmatic, thoughtful and largely depoliticised presentation.
This is very welcome, as it offers hope the NBN review will be evidence-based rather than a political witch hunt.
Matthew Sorell is a Senior Lecturer at the School of Electrical and Electronic Engineering at University of Adelaide. This post originally appeared on The Conversation.
Turnbull made it clear he had asked for the board members' resignations not because he had any particular disagreement, but to give the new government flexibility to appoint a board to meet its new agenda. In fact, he was at pains to recognise the achievements of the NBN board to date.
He stipulated business as usual would continue where work is advanced and contracts are in place. For those of us lucky enough to live in places where the rollout is already happening, this is great news.
Turnbull was at pains to point out that elsewhere, the review may very well demonstrate that fibre-to-the-premises (FTTP) is the best solution, and that his position on technology choice is “truly agnostic”. He mentioned very-high-bit-rate digital subscriber line (VDSL) could be deployed in edge locations where fixed wireless technology is currently being proposed and FTTP is not going to be offered.
But there are some concerns. These really boil down to politicisation not of the technical side, but of interference with project management. Three red flags caught my attention: frequent reference to the NBN not meeting its targets, weekly reporting of progress, and the cost benefit analysis proposed as part of the 60-day review.
Any large infrastructure project, whether building a telecommunications network, a road, or a navy ship, requires extensive management for the project to get delivered on time. It is immensely frustrating that NBN Co in its current form was pressured to provide a timeline of rollout by connections, and that this will continue. A real project does not work this way.
The biggest factor causing the delay to connections was negotiating an arrangement for access with Telstra. The outcome is a faster and much cheaper rollout using existing ducts, but starting later. Meanwhile, priorities were shifted to other parts of the network, such as fixed wireless, satellites, and the transit network.
Committing to a certain number of connections per week at this stage is nonsense, and reporting connections per week is a meaningless exercise in micromanagement.
A cost benefit analysis is a worthwhile exercise, but only to some extent. Such analysis runs a high risk of being biased and, therefore, useless. The notion of “cheaper” and “faster” have not been defined.
There are few precedents on which to base assumptions, and there is a high risk that the review will be skewed towards an outcome thought to be desirable to the minister, or to defend against the perceived desires of the Minister. It was refreshing to hear Mr Turnbull’s strong statement that he wishes to hear the facts, not spin. We will see what happens.
What is meant by “cheaper”? The NBN is an infrastructure project with ongoing expenses. A cost benefit analysis is usually expressed in net present value, essentially a dollar amount at today’s prices.
However, the way in which future value is discounted varies according to the nature of the project. It is not clear whether such a figure should include just the infrastructure or the ongoing maintenance and related operations, nor for how long we wish to consider the project.
It is probably true that the use of VDSL over existing copper lines will be cheaper than FTTP over the next ten years. But fibre infrastructure has a 50-year lifetime. What is the net present value of the entire project over 30 or 50 years?
There are further problems here because the value has to be measured against the contribution of the major shareholder, the government. Is the concern the total amount of money paid from Treasury? If so, the cost benefit analysis will need to quantify productivity and economic growth and the effect this has on taxation revenue.
If the concern is peak funding, this has consequences for other government programs in the short term, but this is something the NBN Co can’t assess as part of its internal cost benefit analysis.
My frank assessment is that we will probably see a lot more fibre being delivered than we were expecting under the new government. DSL technologies on copper will be exploited where these can deliver a better connection than existing arrangements, even if it means these customers wait longer for their inevitable fibre connection.
While there are some concerns around political interference, overall I am encouraged by yesterday’s announcement.