Tuesday, December 31, 2013

Broadband - Looking Back and Looking Forward

As December draws to a close we get to reflect on the year that was.  With an election it was always going to be an eventful political year, but it has been even more eventful when it came to the NBN.

Ultimately it got off to a bad start when in February at Estimates NBN Co retreated from a forecast made to an October 2012 JCNBN hearing that they would pass 300,000 premises by June 2013, advising that they were now only forecasting the original target of 286,000.

Matters became worse when on 21 March they advised they would not meet that target, due to contractors not mobilising resources.

But ultimately everything was upended by Coalition Policy and the election outcome.

Coalition Policy

In September 2012 in an article published on business media outlet Business Spectator, Mr Turnbull wrote that he could not provide precise financial figures regarding the difference between the Coalition’s rival NBN policy and the Government’s existing project:

I have been careful not to nominate a particular sum of money as the difference between what we would do and Labor’s current plan. For a start there is enormous scepticism that the NBN Co project can be completed within the cost and timeframe of their business plan. Several very experienced civil contractors and engineers have said to us recently that they think the actual build cost is likely to be $80 to $100 billion for example.

The then Opposition Leader Mr Abbott in February in a speech to CEDA had said that "If we don’t go ahead with the National Broadband Network in its current form, that’s about $50 billion less that the Commonwealth will need to borrow."

Confusion about the basis for that claim came in April 2013 Mr Turnbull released his own policy and decided to be incredibly precise about his own claims of the cost of the existing NBN plan and about the cost of his alternative. This was the day of the infamous $94 billion claim (which was peak funding).

In defending the policy, Treasurer Joe Hockey even climbed in with a comment on an Alan Kohler piece saying "The fact that our NBN policy saves taxpayers over $60bn is obviously so irrelevant to Alan that it does not rate a reference." 

That $60 billion saving claim was based on the peak funding difference between the Coalition costing of the Labor plan and the Coalition costing of their own.

But the real core of the Coalition policy was the need for three essential reviews - of the cost of the NBN and alternatives, of the availability and quality of broadband and a cost benefit analysis.  These were to be concluded within 60 days, 90 days and six months respectively.

The first observation is that none of these targets have been met.  The Strategic Review "clock" was reset a month till the Board and Management were changed, but at the 60 days all that happened was the Minister got a draft.  It took a further ten days to be released.

The broadband availability and quality report was due on 23 December, but all we got was a short 5 page summary.  It appears that all the Department knows is the location of DSLAMs and the number of available ports. Everything else is derived from that.  We have to wait to an unspecified date for the full detail.

The cost benefit analysis team was only appointed the day the Strategic Review was released and their six months start now. We don't know whether the six months is till the Minister gets his draft or till there is a final for release.

That itself is not a great start.  But the content of the two reviews released thus far doesn't really help either.

Strategic Review

The Strategic Review released on 12 December resulted in Mr Turnbull having to inform the Parliament that he could not meet the goal of providing everyone with a speed of 25 Mbps or better by 2016.

The review itself contained a "Revised Outlook" for the existing project, a "radically redesigned" FTTP rollout and a number of alternatives - with the favoured alternative labelled Multi-Technology Mix - or MTM - or Malcolm Turnbull's mess.

It is not my intention to fully compare these. The review itself is being interrogated by the Senate Select Committee on the National Broadband Network (SSCNBN). 

But I do want to share a few diagrams.  The first compares deployment timelines. 

This is where Mr Turnbull had to admit that the MTM model doesn't get you to 100% with 25 Mbps by 2016.The leap, however, from 22% under the revised outlook to 43% under MTM looks impressive.
However when you examine the previous exhibit you realise this is nowhere near as impressive as it sounds.
Of the 40% of ALL premises able to access 25 Mbps by 2016, more than half are doing so because HFC is suddenly added to the calculation. This is without building any more HFC at this stage. 
Or to convert the percentages from Exhibit 4-2 to a base of the fixed line footprint rather than all premises we see the 43% is made up of 13% FTTP, 5% FTTN and 25% HFC.  Now note that the Revised Outlook has the comparable number of FTTP premises at 22%.  Somehow we are told to believe that passing 18% of premises with new infrastructure instead of 22% is somehow faster.
We then also get to the second issue - the promise of 90% having 50 Mbps by 2019.  Exhibit 4-3 shows they just squeak over this line, but in evidence to the SSCNBN the Executive Chairman was not prepared to guarantee that speed.
But it is also another major deviation from Coalition Policy. The policy document was unequivocal.  Completion would be 2019, and there would be FTTN in all the areas where there isn't FTTP.
But the MTM model is completed a year later and doesn't build where there is HFC - in fact HFC grows to cover a further 7%. 
It is then that the financials become interesting.  Coalition Policy was quite unequivocal when it came to the difference between Labor's policy and theirs.
This is really farcical in hindsight, not least of which is the fact that the Strategic Review uses the same revenue assumptions for all the models - so there is no real "price difference" (There is, however, a quite idiotic approach to argue what prices would need to be to get a 7% return, despite having already cruelled the cashflow by reducing prices).

The two stand-out figures are CapEx of $20.4 billion, and peak funding (unlevered in Coalition case - i.e. as all equity) of $29.5 billion.

How does this stack up with the MTM model?

Cumulative CapEx blows out $10 billion to $30 billion, and peak funding (all equity) hits $39 billion while it is $41 billion if debt is raised (a ten to eleven billion dollar blow-out on Coalition estimates). 

Mr Turnbull was asked on 7:30 about how he was missing his 2016 target, and his plan would cost $11 billion more than he'd said.  He replied;

And the fact is that the NBN Co is a much bigger mess than even we had thought it is. We have got to stop the spin.

But the fact is, it is not in worse shape than he'd claimed NBN Co was in when he launched his policy.  He'd claimed a peak funding of $94 billion and completion by 2025.  The most the Strategic Review could back in was peak funding of $73 billion and completion by 2024. In fact, NBN Co was in better shape than he had claimed.

The single most interesting paragraph in the whole of the Strategic Review comes at the end of Chapter 2 which had detailed the Revised Outlook.  It said;

In other words, the most important thing to do right now is to get over the reviews and the management changes and roll out FTTP as quickly as possible as any other changes are made.


As far back as his speech to the National Press Club in August 2011 Mr Turnbull has made a claim that "The NBN extends high-speed broadband to at least 2 million premises with inadequate service."  That claim when analysed was based on some figures included in the NBN Implementation Study, which in turn was quoting from a 2003 Senate Committee hearing, and even at that hearing the figures were being referred to as already including cases of remediation.

In fact the full footnote to the speech read;

The McKinsey/KPMG NBN Implementation Study (2010) identified 1.2 million pair gain or RIM lines where ADSL was not accessible (p.190), and another 0.4 million premises in the ‘last 7 per cent’ (p.282) where DSLAMs hadn’t been installed in Band 4 exchanges. In addition there is another imprecisely quantified cohort of underserved premises which are in Exchange Service Areas where ADSL2+ is available but the length of the copper run from the DSLAM is too long to allow reasonable speeds.  ADSL2+ has a theoretical maximum download speed of 24 mbps and upload speed of 1 mbps. But average download speeds in Australia are around 10mbps and 63 per cent of users are at least 2km from the exchange (NBN Corporate Plan 2011-13, p. 40). If distance from the exchange exceeds 4km maximum theoretical download speed falls below 6 mbps. FTTN/FTTC resolves this problem by reducing the average and maximum lengths for copper runs.

The "2 million lines" claim featured highly in Coalition statements, including;

  • “When they announced this scheme, when they announced this three or four years ago, there were 2 million Australians that couldn't, on their wire line, download a YouTube video. There still are.” Malcolm Turnbull Lateline 27 Nov 2012)
  • “Much of Australia has good communications infrastructure but not all; at least 2 million premises cannot access broadband or are constrained by limited speeds.” (Malcolm Turnbull website 7 Dec 2011)
  • "If you look at the Implementation Study prepared by KPMG and McKinsey, the study notes that there are about 2 million premises where you can’t get a decent DSL service.” (Paul Fletcher Twisted Wire 20 Dec 2012)
  • “By comparison there's an estimated two million households and businesses that have substandard broadband and don't appear to be getting any level of priority.” (Senator Birmingham to AFR 12 Feb 2013)
  • “It is scarcely believable so little has been achieved since 2007 given there are approximately 2 million Australian households which have sub-­‐standard fixed broadband or no access at all.” (Dr Bill Glasson, LNP candidate for Griffith, letter to Department Secretary to resign as Broadband Champion 22 Feb 2013)
  • "Alan, the tragedy of all this is that there are at least two million households in Australia who don't have - basically don't have broadband at all.” (Malcolm Turnbull, interview with Alan Jones, 27 Feb 2013)

The two million lines suddenly became 1.5 million after the election.  In his Ministerial Statement accompanying the release of the Strategic Review Mr Turnbull said;

Shortly we will publish the first ever analysis of broadband availability in Australia. It will confirm areas encompassing about 1½ million premises have little or no broadband. Many are in regional areas.

Somehow magically an election outcome provided 500,000 premises with broadband that didn't have it before!

In announcing the release of the summary of the broadband availability and quality report, the Minister said;

Key findings of the report are that there are approximately 700,000 premises unable to get access to a fixed broadband service and an additional 920,000 premises in areas with estimated median peak download speeds of less than 4.8 megabits per second (Mbps).

So now we have the availability and quality report defining the number with "little or no broadband" as 1.62 million.

But what should we make of this observation?

The first is that the 700,000 with nothing are predominantly, but not exclusively, the areas receiving fixed wireless and satellite.  There appears to be no difference in the approach to these, scheduled for completion in 2015.

More surprising is the claim that a "median peak download speed" of 4.8 Mbps is now the definition of "inadequate".

This dramatically contrasts with a statement Mr Turnbull made as Opposition Leader.  Speaking on radio 4RO on 15 April 2009 Mr Turnbull said:
I've got a Next-G wireless card in my laptop and when I was in Mackay yesterday I was getting 3.5 Mbps … that allows me to do everything I need to do.
In 2009 3.5 Mbps allowed him to "do everything I need" - but by 2013 having less than 4.8 Mbps qualifies as inadequate?

I have no real difficulty with Mr Turnbull over four years realising that requirements of download speeds are increasing.  It's just that his own policy doesn't. 

There are many who would argue that every premise without FTTP has inadequate broadband, the real number is 12 million not 1.5 or 2 million.

The other challenge revealed in the report is that "approximately 1.4 million premises (13 per cent) are in areas where fewer than 40 per cent of premises can access a fixed broadband service." These are the areas from which the majority of complaints come - they are the complaints about there being "no port available".  This is an issue that should be fixable. Presumably for exchange based DSLAMs as FTTP gets rolled out there will be DSLAMs that could be relocated from decommissioned exchanges.  For RIMs there is a useful conversation to be had with Telstra about a further extension of the top-hat program.

Where to from here

The Senate Select Committee will continue its evaluation of the Strategic Review.  Irrespective of that the review itself for the MTM scenario is predicated on continuing to rebuild momentum in the FTTP deployment.  The assumptions for the MTM model (page 96) include that FTTN deployment will commence in the second half of CY15.  That is vastly different from Dr Switkowski's expectation in evidence of getting FTTN rollout happening by the end of CY14.

Everyone who wants FTTP needs to focus less on what the alternatives might be, and more on encouraging the Minister to instruct NBN Co to build the deployment rate as fast as they can while they undertake any redesign.  The MTM model forecasts 26% FTTP which is higher than the 21% in the Coalition Policy. 

The best way to do this is by writing to local MPs. Ideally clever people inside the various NBN campaigns could construct a simple website that uses he AECs digitised maps to do an electorate lookup for an address, then prepare a completed and addressed draft letter or e-mail on the theme of encouraging the Minister to instruct NBN Co to continue to rollout FTTP as quickly as possible while it continues to undertake its review activities.  Lots and lots of letters, especially ones that are slightly tweeked by the individual help.

The second thing is to focus more on why the higher speeds and a four port NTD are needed for the applications that we expect will be common once the NBN becomes universally available. 

This is an area where people and organisations should use the opportunity to make submissions to the Senate Select Committee (by 31 January). The Committee has limited direct resources of its own, so don't expect that they will be able to find something just because it has been previously published.  Submissions from people already using fibre, or people with plans to use fibre either for themselves or in applications will help make the case.

The future of broadband in Australia is still a live policy question.  To get a good outcome requires good people to be very active.

1 comment:

Sydney said...

As usual David analytically exposes relevant facts and figures that are of serious importance to understand the complexities of the proposed NBN roll-out. His exposure requires detailed study and serious respect by all.