Sydney, 18 November 2013
I’m pleased to be here this morning to help kick off the NBN Rebooted Conference, and I congratulate Michael Lee and John Stanton at the Communications Alliance and Grahame Lynch at Commsday for making this event happen. Likewise, I thank Alcatel Lucent, the major sponsor, and the rest of the sponsors for their support.
It’s tremendously encouraging that people from every corner of the telecommunications industry have chosen to attend this event in such numbers.
From the outset, the National Broadband Network has been a colossal undertaking, well beyond the capabilities of any single firm or any government to execute on their own.
I’ve previously expressed the view that the Government should not be nearly as involved in this project as we are.
Elsewhere in the world the standard model of building new generation networks has been for governments to provide subsidies to support deployment in less economic, typically rural and remote, areas and for the project and business execution risk to be carried by those best able to manage it – the private sector.
However the previous Labor Government turned all that on its head. Instead of Government having a certain financial commitment, it was the private telcos who received financial certainty in the form of a cheque. And all of the execution risk of the largest, most complex, broadband deployment in the world was assumed by the Australian Government in the form of a start up. The folly remains breathtaking to this day and more obvious every day.
We would never, ever have undertaken this project in this manner. And you only have to look to New Zealand or indeed the UK to see much wiser and more cost effective approaches.
However we are where we are.
But make no mistake; to get this project back on track and completed in a reasonable timeframe at a reasonable cost is a daunting challenge.
And it is abundantly clear to me that to achieve that goal, we are going to need commitment and flexibility, patience and hard work from the entire industry – not just from Telstra and Optus, not just from iiNet and Vodafone, but from across the whole sector: service providers, carriers, vendors, contractors, policymakers and regulators.
Over the next two days you’ll have the opportunity to hear voices representing many of these points of view.
So will my colleagues and I – and let me assure you that we’ll be listening carefully.
The profoundly disappointing progress on the NBN so far holds many lessons. But one of the most important is that Government simply doesn’t possess all the answers, especially when it comes to executing a complex plan in a fast-changing industry such as ours.
On the contrary, the woeful performance of the NBN rollout against its financial and operational targets so far indicates that there is a vast amount that the Government and NBN Co have to learn. It’s my hope that this event can make a constructive contribution to that exercise.
I’ll speak in greater detail about the current state of play at NBN Co shortly, to the extent that I’m able to do so without pre-judging or pre-empting the findings of the Strategic Review presently underway and due to be presented by the board of NBN Co to the Government in early December.
But first let me make a few points about the broadband policy the Coalition took to the election on 7 September, and some of its implications for the way the Government is going to approach the NBN.
You’ve heard me say many times that our main objections to the previous Government’s NBN were its excessive cost, the significant length of time it would take to complete, and its shifting of a very large amount of risk from private investors to taxpayers and the public purse. Those all remain valid concerns.
But today I want to elaborate on another of our core issues with the current plan – the more subtle but hardly less important objection that it unnecessarily forecloses on optionality, by setting in concrete choices that can valuably be deferred.
History tells us that policy-makers and politicians are not very good at predicting what’s going to happen next and there is not shortage of evidence for that in the telecommunications and technology industries.
In hindsight, of course, ministers and senior public servants claim to have seen the signs of an unexpected development even as they scramble to put together a retrospective rationale. But the truth is that governments are all too often blindsided, particularly when a shift is distinct from previous experience.
Nicholas Taleb describes the framework that we should ideally bring to forecasting:
“Probability is not a mere computation of odds on the dice or more complicated variants; it is the acceptance of the lack of certainty in our knowledge and the development of methods for dealing with our ignorance.”
But that is easier said than done. And in the case of the ICT industries generally, and the NBN specifically, the dangers are magnified. We can see this in the rapidly changing external context for the NBN that have occurred since April 2009, when the NBN Mark 1 was ditched and the current one-technology-fits-93% FTTP plan proposed.
At the time NBN Mark 2 was unveiled, Cisco’s first VNI index, which had been released in the middle of 2008 predicted consumer use of Internet traffic (in terms of volume) would grow at an annual rate of 49 per cent over the half-decade from 2007 to 2012 .
As it happens, that prediction turned out to be extraordinarily accurate – consumer Internet volumes grew at a rate very close to Cisco’s estimate over those five years.
Today, in contrast, we see that the most recent VNI, which was released in June 2013, forecasts a growth rate for total consumer use of data over the five years from 2012 to 2017 of just 23 per cent. Rather than data volumes doubling every two years, now they are doubling every four years.
That is still very rapid growth, but not nearly as intimidating to carriers as it was.
And even more importantly, large growths in the volume of data being carried across the Internet and indeed being consumed by households are driven overwhelmingly by video consumption and it is self evident that data consumption can continue to grow without a correlated growth in line speeds.
Granular analyses of current and projected broadband consumption falls short of the very high predicted line speed requirements which all too often assume there is a linear correlation between volume growth and the demand for greater line speed .
Plainly, if costly and irreversible investment decisions can be deferred until demand either materializes or can reasonably be foreseen, savings arising from the time value of money are not the only economic value generated. There is also very real value in keeping the option open of doing something different than what might seem at a certain point in time to be the obvious answer – of responding to changing technology and changing market conditions.
Changing Market Conditions
Likewise, in the past five years we have seen an explosion in new techniques for wringing higher data rate broadband out of copper telephone wires.
In fact between the widespread deployment of fibre to the cabinet plus VDSL2, the emergence of vectoring and bonding, the initial commercial trialing of fibre to the distribution point, and emerging protocols such as g.fast, it is fair to describe the past five years as having encompassed a revolution in the way copper is used.
Equally important changes are continuing to come at us at a rapid pace.
Wireless networks are increasingly being rolled out using small cell technologies that dramatically lower the cost of extending the reach and throughput of this channel – but also creating the need for deep fibre backhaul networks in urban areas.
Radiofrequency spectrum is used more efficiently, greatly increasing the data that can be sent over wireless networks. And mobility has proven to be a key source of productivity and convenience gains from more advanced communications, even as larger-screen higher-definition TVs (the key driver of consumer use of bandwidth) give way for many of us to personal TV viewing on tablets.
A more nimble approach to designing (and imagining) the NBN can preserve a great deal of this optionality.
So next time you read or hear someone claim that FTTN is a ‘waste’ because the network will just have to be upgraded to FTTP some time later pause for a moment and consider optionality. Consider its value, and consider the vast changes we’ve seen in so many areas since 2009.
And consider what a difference they have made to the economic and technological parameters within which the NBN will be rolled out.
The previous Government didn’t get this point. It was naively willing to forego any optionality whatsoever in its reckless pursuit of what was seen to be a ‘pure’ technological solution that cut through three knotty problems at a stroke – very fast broadband, improvements to regional services, and the unwinding of Telstra’s vertical integration.
The resulting plan for the NBN was portrayed to taxpayers and consumers as commercially viable and free of risk. Heroic revenue assumptions with unprecedented growths in ARPU marched out into the future all the way to an equally heroic assumed sale in 2040 based on an eye-popping EBIT margin and multiple to produce a huge terminal value. And as if this was not enough, it was then discounted back to today’s money at a discount rate appropriate to a mature, low risk venture .
I see The Australian has reported one investment bank told the previous Government that in its view the project had a negative NPV of $31 billion – based it would seem on a more conservative view of both the amount and riskiness of the future cashflows. Senator Penny Wong has complained that this advice is a “selective” account of the total received by Labor in Government. Well there is a simple solution to that. Mr Shorten, as Opposition Leader, can authorize the release of cabinet papers of the previous Government, otherwise inaccessible to the Coalition.
He should do so.
Where is the NBN up to?
One result of the glacial rollout of the existing monolithic approach to the NBN is that Australia has continued to fall in rank on most indices of broadband quality – we are a troubling 29th in fixed broadband penetration, according to the 2013 United Nations ‘State of Broadband Report’ . Last year we ranked 26th .
Just as the NBN volume rollout was planned to be ramping up, we are sliding down the league tables of international competitiveness in terms of broadband.
The NBN rollout is simply not reaching enough Australian households and businesses in a timeframe or at a cost that is sustainable. If you want to see that, check the weekly data we now publish on the NBN Co website.
Australians consumers are paying a high price for the rigidity of the previous plan, which Minister Conroy was unwilling to admit until after the voters had had their say.
As the Strategic Review delves further into the current state of the rollout, we are slowly getting a better handle on what NBN Co really was capable of achieving under the outgoing management, and how we need to bolster its capabilities.
Shortly before the election, the NBN Co revised its June 30 2014 premises passed target down to 600,000 brownfield premises, about half what had been estimated in August the previous year.
However it appears that even that estimate was unduly optimistic and it is likely that when new, realistic and genuinely achievable targets are set they will be materially lower – around 450,000. Compare that to Labor’s December 2010 target of 2.2 million brownfield premises passed with fibre by June 2014, and we see NBN Co is on a path to achieve only 21 per cent of the original forecast.
We can’t change the past, but we’ve moved quickly since the election to make changes at NBN Co that equip it to transition to a more predictable, affordable and achievable rollout.
Critically, we’ve advised NBN Co that the network should be rolled out cost-effectively using the technology best matched to the circumstances of each locality. This is fundamentally different to the Conroy model of decreeing GPON for 93 per cent of the country.
To this end, on September 24 the Government provided NBN Co with an interim Statement of Expectations to guide construction and operational activities while a series of reviews of the NBN project are carried out.
In the transition period, NBN Co aims to avoid service disruption for consumers; minimise the impact on the construction industry of the switch to a mixed technology rollout by continuing FTTP deployment where building contracts are in place and also in areas where fibre is the appropriate technology such as business precincts and greenfield developments; and achieve a less costly and quicker rollout plan as seamlessly as possible.
On October 3 we announced the terms of reference of the strategic review and changes to the Board and senior management, including the appointment of Dr Ziggy Switkowski as Executive Chair of NBN Co.
On November 12 we appointed three further directors – Justin Milne, Patrick Flannigan and Simon Hackett – with many decades of combined industry experience to the NBN Co Board. Several key changes in the executive ranks have also been made, notably the appointments of Greg Adcock and JB Rousselot, both experienced former Telstra executives. The search for a CEO is well advanced as well.
The above reforms are a good platform to start sorting out the mess. The next step is the Strategic Review, which will provide a frank assessment of the trade-offs between time to market, service levels and costs involved in delivering high-speed broadband to all Australians. This Review must be an exercise in truth telling. We want hand on heart true, realistic and achievable options prudently costed and scoped on which we can make weighty decisions. There is no longer any room at NBN Co for spin or for telling the Minister what people imagine he wants to hear.
In short I want the team, the management, the board at NBN Co to regard every forecast, every decision as something they would be prepared to defend in a prospectus for a public listed company. It is critical for the health of that company that it has forecasts and goals which it can meet. I can only imagine how debilitating it must have been over the last three years to be constantly setting and missing and setting and missing one forecast after another.
There is no pre-determined outcome. I have always said that everything is on the table. The Government will be guided by evidence-based recommendations, whatever they might be. That is surely the only logical position a responsible Government could take.
The Strategic Review will give NBN Co and the Government options, and a clearer fact base upon which to base our decisions. Of course the process of examining, investigating and revising the business case for the NBN will not end with the Review – it should also be seen as a stepping stone to a new Corporate plan in mid-2014.
In closing, let me remind you that up to two million households and businesses across Australia cannot get basic fixed-line broadband at present. Addressing these underserviced areas first is a key objective of our NBN policy.
As a priority my Department, with the assistance of NBN Co and private carriers, will provide the Government and Parliament with a ranking of broadband quality and availability in all areas of Australia. This ranking will be published for comment and review and guide future prioritisation of the rollout.
At long last, the NBN rollout will be directed towards the areas that need it the most.
Some renegotiation of the Telstra Definitive Agreements is required and we look forward to concluding that before too long and in a spirit of collaboration and partnership. Telstra after all is NBN Co’s biggest customer and has a considerable vested interest in the project’s successful completion.
None of this will be easy, however, and I expect we will suffer occasional setbacks along the way – you can’t expect a turnaround of this magnitude to unfold without a hitch. But I am cautiously optimistic about the NBN, for the first time in its four and a half years of existence. We are assembling the right team and the Strategic Review will help us frame the right plan. But please remember what I said at the outset – building an NBN that works isn’t a matter for Government alone. I need your help, your expertise and wisdom, your advice, and your patience.
Thank you for your time this morning.
 Cisco VNI – Projected Five Year Growth Rate, Bandwidth Usage:
 See for example Robert Kenny & Tom Broughton “Domestic Demand for Bandwidth” 5 November 2013 produced for the UK Broadband Stakeholder Group.
 The terminal value in 2040 for NBN Co required to generate a 7.1 per cent return on equity will be a matter examined by the Strategic Review. Most publicly available independent estimates have suggested the required disposal value of NBN Co to achieve the benchmark is in the order of $90 billion, or 10-12 times EBITDA – a very aggressive assumption.
 Available online here, p.92
 Available online here, p.84