NBN Co tied up in fibre knots

The past couple of weeks have not been pleasant for NBN Co and the latest appearance by NBN Co CEO Bill Morrow, COO Greg Adcock and CFO Stephen Rue at the Senate Select Committee on the National Broadband Network only compounded the problem.

Morrow must have been particularly discomfited by the proceedings given his suggestions of a conspiracy by disgruntled former employees or “FTTP zealots” hell-bent on making life difficult for the new NBN Co regime. Is there really a plan afoot to fabricate and release NBN Co FTTP trial reports to undermine the government’s multi-technology mix (MTM) NBN? Well, no.

So what’s really happening over at NBN Co and why has the NBN Co management team managed to notch up such a bad fortnight?

Good news can be bad news

An important characteristic of telecommunication infrastructure rollouts, especially when new technologies are first deployed, is the scope for subtle yet significant improvements to the rollout implementation and technologies being used.

Improvements that save time and cost or improve reliability, resiliency and sustainability are commonplace when relatively new technologies are being used. Teams that supply the equipment, plan and design or construct the network work together to not only reduce installation time and costs but also cut life cycle costs and improve system performance.

So it shouldn’t be unusual for NBN Co to promote the achievements of the people working to complete the largest infrastructure project in Australia’s history. That is, unless the achievements don’t support the government’s MTM NBN.

Do you think that NBN Co will bury achievements that show that a national fibre-to-the-premises (FTTP) rollout could be completed within a couple of years of the limited fibre-to-the-node (FTTN) rollout or reduce the total cost of ownership for a national FTTP network for the first ten years of operation to less than what it will cost for a national FTTN network?

One would certainly hope not.

The Melton moment

Last month Fairfax reported the confidential leaked results of a FTTP rollout pilot study completed in Melton, Victoria that utilised an updated approach that incorporated aspects of the “radically redesigned” FTTP option described in Section 3.3 of NBN Co’s Strategic Review Report (PDF).

The changes made to the scheduled deployment of FTTP to 2484 premises reduced the time for the rollout from the average of 344 days in other areas to about 104 days in Melton. Fairfax reported that the confidential report stated that “ninety per cent of buildings were serviceable by fibre by the end of August - 61 per cent faster and 50 per cent more cheaply than in areas using previous rollout models.”

Reporting of the Melton FTTP trial results continued at ZDNet but by this time NBN Co had issued a strong response to the story. Last Friday, Senator Stephen Conroy repeatedly questioned Morrow at the Senate select committee about NBN Co’s strong response and Morrow told the committee that “the immediate response from NBN Co’s media team could have been better.”

Contained in NBN Co’s response were the assertions that the document leaked to the media had not been peer-reviewed and still needed to undergo an internal validation process prior to being presented to senior management for consideration.

At the Senate select committee Adcock raised doubts about the figures included in the document especially the 104 days taken to complete the rollout and he went on to state that “early indications are that the fault rate in this FSAM are much higher than the national average.”

On the one hand we’re told that the results in the document have not been validated and peer-reviewed and should therefore be dismissed until such time as a review can be carried out yet Adcock felt there was a need to make an unsubstantiated assertion to the committee about a higher than average fault rate.

Melton was a trial and in any trial there will always be things that go well and some that don’t. Possibly the higher than average fault rate, claimed by Adcock, was due to the rollout teams being unfamiliar with the new rollout approach. It should be anticipated that as the rollout team becomes more familiar with the new rollout approach the fault rate would subside and eventually fall below the current average fault rate.

Rollout costs versus life cycle costs

The key reason that NBN Co is sticking with a FTTN is the government’s Statement of Expectations (PDF) for the NBN that requires NBN Co to “determine which technologies are utilised on an area-by-area basis so as to minimise peak funding, optimise economic returns and enhance the Company’s viability.”

This requirement has been interpreted by NBN Co to mean that it will do “whatever it takes to get the NBN to you as quickly as possible with the least cost and the least disruption while meeting your technology needs.”

And herein lies the problem. It was hoped that the NBN Cost Benefit Analysis (CBA) would shed some light on the actual spend for the alternate NBN technologies but the CBA is of no reasonable value in the current debate.

What’s needed now is a Life Cycle Cost and Performance Analysis (LCCPA), a process that should have been completed in 2010 and updated regularly since then. LCCPA are compulsory for major projects undertaken by government departments including Defence, so why has NBN Co escaped this requirement? As a government authority it appears to be at arm’s length from the mandatory requirements that have been placed on government departments, including Defence.

Let me give you an example of what a LCCPA would show. The cost of FTTP connections started out at around $2000 and has slowly dropped to a figure today of about $1200 per connection. We don’t know the exact figure today because NBN Co often redacts the connection cost in documents released to the public, an example of this can be found in Section 3.3 of the NBN Strategic Review.

The government signed a $150 million FTTN deal with Telstra in July 2014 to rollout 1000 nodes to connect about 206,000 premises in New South Wales and Queensland to the NBN. What this means is that FTTN connections will cost about $730 in this pilot.

So currently there’s an upfront cost differential of about $470 per connection between FTTN and FTTP and funds borrowed to meet the extra cost of FTTP would need to be repaid including interest.

NBN Co recently stated that FTTN cabinet power will cost about $2000 per year, a cost that does not exist for FTTP. The FTTN cabinets will be placed near existing telecommunication pillars which contain connections to existing copper services for about 200 premises. This means the power cost per connection is about $10 per year and when inflation is taken into account this figure becomes $13.70 after 10 years and $27.20 after 30 years.

And a BIS Shrapnel report published in August 2012 stated that “the technological superiority of optic fibre over Australia's ageing fixed-line copper network is estimated to reduce industry maintenance costs by between AU$600 million to AU$700 million per annum once fully deployed.”

This means is that the reduction in infrastructure maintenance costs if FTTP is rolled out will be about $35 per connection per year by 2021.

Now if we start in 2015 and add $30 per year for maintenance with $10 for power and increase by inflation over twelve years the result is about $584. So by about 2027 the extra cost of rolling out FTTP would be repaid in full including interest using the maintenance and power savings alone and from this point forward there would be an annual saving of about $40 per connection in current dollars.

And this simple analysis does not include the FTTN cabinet batteries ($3000 per cabinet), the higher cost of FTTN equipment, the cost of dealing with water ingress which will have an even greater effect on VDSL2 with vectoring than it has on ADSL2 , the cost of upgrading the HFC to DOCSIS 3.1 in five years, the cost of upgrading the FTTN to G.Fast in five years and so on.

NBN Co’s claim, reported by ZDNet, that the $120 million operating cost to power the 60,000 FTTN cabinets across Australia each year will be less than the up-front capital cost of rolling out FTTP is meaningless as there are many other costs to take into account.

NBN Co must carry out a LCCPA to provide a complete and detailed accounting of the technology, operational and life cycle costs and there is no evidence available that NBN Co has carried out a LCCPA other than to make rough estimates that would not be acceptable if the NBN was being built by a government department.

Alternative approaches

The current rollout cost of FTTP is about $500 per connection more than FTTN. Yes that’s right $500, which is less than the cost of an iPhone or a Samsung Galaxy.

If the government is so concerned about $500 per household it should ask NBN Co to consider alternate approaches to recoup the $500. Two approaches that could be adopted include a $500 connection fee in more affluent inner urban areas (means tested) or community working parties to carry out the installation of fibre from pits into premises which is an approach that was used in Northern Europe.

Or the government could wake up to the fact that the whole NBN debate is ridiculous. Why proceed with a flawed policy when the far superior future proof solution will only cost an extra $500 per household? That is $20 billion for a national FTTP rollout spread over 10 years compared with the $80 billion for new road funding spread over four years agreed between the Commonwealth and state governments at the last budget.

The pressure will only grow

NBN Co senior management will face increasing pressure as time moves forward because it’s impossible to defend an untenable position. The leaks will continue from NBN Co and in time the leaks will become a flood, because there will always be those that believe the truth must be told. In an effort to plug the leaks NBN Co risks becoming less transparent than at any time in the last five years.

There are no “FTTP zealots” in Australia just concerned citizens that want NBN Co to build a NBN that is cost effective, has a life-time of 30-50 years and provides an operating performance that will ensure Australia is a leader in the global digital economy.

Morrow needs to consider what has happened over the past fortnight and realise that it’s time for NBN Co to change tack and embrace the debate by increasing transparency and providing guidance that can be supported by detailed substantiated analysis and logical argument.

Mark Gregory is a Senior Lecturer in the School of Electrical and Computer Engineering at RMIT University