Communications minister Stephen Conroy has had a rough year. He has been suffering from chronic knee pain and recently suffered a bout of pneumonia.

Admirably, he hasn’t let these issues affect his work rate as a minister, still making good on his commitments despite obvious personal discomfort. Of course, things haven’t been too flash on the political front for the government either and as the number 4 ranking member of that government, Conroy wouldn’t be finding much on the work front to cheer him up from his health issues.

In this context, it wasn’t surprising then to see him go on the attack from the onset of yesterday’s NBN corporate plan press conference. Although he could have chosen to focus on the reformist impulses behind the NBN he, instead, decided to attack those he saw as misrepresenting the facts about it.

Various entities including Malcolm Turnbull, Henry Ergas and the Australian Financial Review all copped a serve: some of the criticisms fair, others such as that of Ergas not so fair. Ergas’ early 2009 calculations of a $120-$200 overall retail ARPU were based on an NBN earning a 15% ROI and a preservation of existing retail margins. It is noteworthy that NBN Co itself predicts its own wholesale ARPUs to eventually hit $100 at around the same time it hopes to pay for itself. It is also noteworthy that retailers will be taking margins hits in their NBN offers compared to DSL services based on unbundled copper, just as the previous NBN corporate plan forecasted.

The accompanying press release also made its own inaccurate comments, suggesting that the NBN’s total capital expenditure was still much less than the $43 billion originally announced in 2009. But that original figure was never framed as just the capital spend: it was described as the amount to be “invested” in the network by government and the private sector. The total investment requirement for the NBN from government and private sector-sourced equity and debt as of yesterday? $44.1b. This is not “significantly less than the $43 billion originally announced.”

The fact that the actual corporate plan itself wasn’t released until 30 minutes into the press conference gave the act away—also many of its key metrics were published on page 76 of the actual document! What did the government choose to highlight yesterday? A slight and meaningless revision of the internal return rate by 0.1% over 30 years.

What did it bury? Let’s see: a revised forecast which sees nearly 80% less connections to the NBN by June 2013 than it predicted a mere 20 months ago. A blowout of 8% or $3.2 billion in peak funding requirements. A blowout of over 10% or $2.9 billion in the government equity requirement. And the fact that with nearly $3 billion spent to date, the company will only collect $120m in revenues next financial year and won’t crack $1 billion a year until FY2016.

Rob Oakeshott is right: there isn’t really a great desire for informed scrutiny of the government over the NBN even as its shareholder minister attacks others for their allegedly misinformed scrutiny.

Much of the new corporate plan is structured around the assumption that a greater percentage of NBN users will subscribe to the higher speed (100Mbps and more) services than was originally thought to be the case. There are two justifications for this: one of the experience of the large percentage of early adopters who are subscribing to the higher speed service, the other is the planned migration of nearly 500,000 Optus customers from a 100Mbps DOCSIS 3 HFC network to the NBN.

I’m not sure either justification holds up: how many of the Optus customers have actually elected to pay a premium for top tier service on HFC as opposed to being a passive party to a network upgrade? And do early adopters who are motivated to take an NBN subscription from day one really reflect the broad mass of users who will only take to the NBN when their copper is switched off?

NBN Co seems to have a foot in both camps. Its speed demand projections posit that by 2028 about 40% of its users will remain on the entry level 12/1Mbps plans and about the same—40% or so—will be taking 250Mbps, 500Mbps and 1Gbps services. Only 20% will elect to be on 25Mbps, 50Mbps and 100Mbps services.

This is a curious demand dichotomy if one thinks through its implications on the types of mass market applications, content and services envisaged for the NBN-enabled digital economy.

Despite earlier talk that there had been second thoughts about the contentious Connectivity Virtual Circuit charge, essentially a usage charge designed to pump up ARPUs from averaged and loss-leading access fees, it looks here to stay. NBN Co is offering the sweetener of a CVC charge waiver until customer numbers in each fibre area reach 30,000: elegant given the large impost it levies on RSPs at the early point of the NBN transition when they also must wear legacy copper charges. That said, NBN Co was keen to point out its $20 Mb CVC charge was less than Telstra’s equivalent $33 charge for its existing resold wholesale DSL service. Of course, Telstra aggregates these charges at six state POIs, compared to the 121 envisaged by the NBN.

The elephant in the room yesterday at NBN Co’s corporate headquarters was the pending Federal election in 2013, which many senior members of the government expect to lose. If the government sees out its full term, the FY projections for the NBN suggest that there may only be a couple of hundred thousand premises or so connected to the network by election day.

By that stage, the government will have ploughed the best part of $10 billion into the network. Rightly or wrongly, the project is set up for caricature as another pink batts or school halls scandal.

The peak years for NBN construction at the moment appear to be 2015 and 2016—that is when the bulk of the spend is projected to occur.

I suspect the politics of popular support for such spend, especially as interest bills begin to bite into the surplus or deficit, might have changed, for the negative, by then. As much as many in the industry genuinely want Abbott and co to change their minds and support the NBN in its current incarnation, I’m not even sure the ALP will be so enthusiastic by that stage.

I note the words Julia Gillard used this week to attack what she described as over-investment in power capacity by utilities.

“One quarter of all retail electricity costs – more than $500 a year for a typical family – is spent to meet the costs of peak events that last for less than two days each year in total… It’s like building a ten lane freeway – but with two lanes that are only used or needed for one long weekend. It’s a very good thing that more Australians can afford air conditioners to cool their families on our hottest days… It’s a very bad thing that the supply side response to this is so deeply costly and inefficient, and it’s a clear argument for reform to go further…My challenge, to industry, regulators and state governments, is this: your job now is to ensure that you respond to this with efficient investments. Investment which gets the balance right between affordability and reliability….”

Substitute the words “broadband” for “electricity” and “computers” for “air conditioners” here and it is easy to see the attack lines that will be used against any politician presiding over the current NBN plan over the next few years as it reaches peak spend ahead of meaningful uptake.

Even the most ardent NBN advocates agree that the network is way over-provisioned for current demand and is built with the requirements of the next three to four decades in mind. As iiNet posited on its own corporate blog last week, some 3% of total users likely account for 50% of bandwidth demand. NBN advocacy often seems to confuse the difference between an average and a median (indeed, yesterday’s corporate plan does so in relation to data usage): something a politician who wants to win 50.1% of the two party preferred vote and see a contentious project through should never do.

Get ready for NBN scaleback no matter what happens in terms of government, especially given the bulk of the overall NBN cross subsidy (plus the proposed TUSMA funding) goes to rusted-on voters in safe regional and rural seats—unless Julia Gillard is suffering from massive cognitive dissonance, her own rhetoric on power costs is a sign of where the NBN debate is heading.

Grahame Lynch