NBN cost-benefit analyses are so 2011

NBN cost-benefit analyses are so 2011

Summary: A year ago, Malcolm Turnbull repeatedly proclaimed that a coalition government would stop the NBN in its tracks and conduct a cost-benefit analysis on the NBN. These days, the analysis doesn't even get a look in. Is the idea dead forever — or could it still offer some value?


It wasn't too long ago that one of the rallying cries of the Coalition's NBN opposition was its insistence that Labor conduct a full cost-benefit analysis, comparing the government's fibre-to-the-premise (FttP) strategy with alternatives such as fibre to the node (FttN) and wireless.

By avoiding a formal cost-benefit analysis, Malcolm Turnbull argued over and over, the Rudd and Gillard Labor governments had broken the party's promise for greater transparency on infrastructure projects — and pushed the country along a precipitous course towards what he has repeatedly lambasted as a policy disaster of the worst imaginable sort.

Malcolm Turnbull once promised to subject Labor's NBN to a cost-benefit analysis. But have we now passed the point where it could be useful? (Image by US Navy, public domain)

On the first point, he is correct: Labor certainly seems to have forgotten its commitment to subject major infrastructure spending to the cold calculations of Infrastructure Australia, which seems to have become a cheerleader for infrastructure investment, rather than an independent assessor of it; a post last year by the Institute of Public Affairs described the body as "all but derailed".

The government seemingly remains convinced that a cost-benefit analysis would be inappropriate (Turnbull may prefer the word "inconvenient"), because the current plan does not require the comparative evaluation of two possible uses of public funds.

This is not an entirely baseless conclusion, given the explicit approval of the government's treatment of the NBN as an "off-budget" investment; if we are to take this treatment as gospel, it would be incorrect to position a cost-benefit analysis comparing FttN with FttP, because only the latter involves the direct spending of public monies; the latter is, the government has continually asserted, an investment. Each is a different type of spend, with different parameters and assessable risks.

Whether you buy that argument or not, the Coalition hasn't been talking about cost-benefit analyses much recently. It was just a year ago, remember, that Turnbull was promising that the first act of a Tony Abbott government would be to halt the NBN in its tracks and embark upon a comprehensive cost-benefit analysis to prove once and for all what was the better alternative.

It was just a year ago that Turnbull was promising that the first act of a Tony Abbott government would be to halt the NBN in its tracks.

These days, Turnbull seems to be more pragmatic. He recently admitted that the Coalition would likely leave most of the NBN intact, and has not been repeating — in public, at least — his previous determination to put his foot on NBN Co's proverbial neck. I contacted his office to ask about the status of the cost-benefit analysis, but they declined to comment. However, he didn't mention the cost-benefit analysis at all in his latest major industry speech.

Little surprise that he's speaking more softly; as Mike Quigley said recently, you don't just stop a 2000-person company overnight; there are revenues to be earned, infrastructure to be rolled out and people to be paid. There is also an entire industry that has steadily adapted itself to life in the NBN world — and none of these stakeholders would benefit from, or take kindly to, the kind of arbitrary shutdown to which Turnbull had previously been so committed.

But what of the cost-benefit analysis? Surely, Turnbull's response to the budget revelation that stopping the NBN would incur $1.8 billion in cancellation penalties confirms that even he has conceded that a cost-benefit analysis would favour a continuation of the project, at this point. Labor's NBN now has so much momentum that it seems almost inconceivable for any analysis to find that it would be better to throw the project into chaos and embark on an entirely different course of action in the blind hope of saving a few billion dollars.

If it's elected next year, will the Coalition still run a "we-told-you-so" retrospective cost-benefit analysis?

Does that mean a cost-benefit analysis would be pointless? This is something the Coalition will have to consider if it's elected next year: would it spend time running a cost-benefit analysis on a project that's already in progress? Would the terms of such an analysis potentially include the option of shutting down the NBN, as Turnbull and Tony Abbott had so often promised? Could the updated NBN business plan provide more fuel for the Coalition to argue that a cost-benefit analysis is more relevant than ever? And, if it's elected next year, will the Coalition still run a "we-told-you-so" retrospective cost-benefit analysis in the hopes of securing the moral authority it has claimed for so long?

You'd think so, although it might come off as petty and wasteful for the Coalition to push for such a seemingly pointless cost-benefit analysis out of spite; the black-and-white FttN versus FttP debate became moot some time ago. And yet, there might be residual value in a cost-benefit analysis that compares the two options now on the table — continuing FttP as per Labor's plan, versus keeping what NBN Co has rolled out or committed to up until the election, and injecting FttN into the mix in areas where ADSL2+ actually works well.

This would, if nothing else, put some real weight behind the Coalition's still-rhetorical argument that its plan will be cheaper and faster to implement. Of course, the second option would also have to factor in to the presumably significant costs of renegotiating access arrangements with Telstra, or even of purchasing the copper network outright. And that cost, we may read from Turnbull's recent silence on the cost-benefit analysis question, may have turned out to be the thing that killed the prospect of the analysis for good.

Update: subsequent to the publication of this column, Malcolm Turnbull's office has advised that the Coalition is still committed to an NBN CBA, to be conducted by the Productivity Commission.

What do you think? Why are we not hearing about cost-benefit analyses anymore? Could the Coalition still conduct a meaningful cost-benefit analysis if it were elected next year? Or is there simply no practical point in discussing alternatives anymore?

Updated at 4.49pm: added comment from Malcolm Turnbull's office.

Topics: Broadband, Banking, Government, Government AU, NBN, IT Employment


Australia’s first-world economy relies on first-rate IT and telecommunications innovation. David Braue, an award-winning IT journalist and former Macworld editor, covers its challenges, successes and lessons learned as it uses ICT to assert its leadership in the developing Asia-Pacific region – and strengthen its reputation on the world stage.

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • The coalition has done nothing else but keep changing their view over the last 2 years.

    -first it was "there is nothing wrong with the Internet we currently have and this is a white elephant". Let the private industry do it themselves.

    -"wireless is the panacea and wired Internet is dead". We should do wireless everywhere.

    -"we do agree faster broadband is need but FTTH Is a waste, it should be FTTTN"

    Etc etc etc....

    It is clear their goal posts are CONSTANTLY moving and they make things up along the way based on what the polls are saying.

    At least the NBN seems to have fairly clear goals and is on the ground finally running. The libs seem to still be all chatter and zero solid policy so if they halted construction it would be years before they restart whatever hodge podge network they decide on.
    • The Liberals have no idea what to do and would just go back to the "do nothing" policy we had under Howard, Alston and Coonan.
      • If a CBA had been done on the Opera House (started by a Labor Gov) it wouldn't exist. Had a CBA been done on the Internet itself, would the Internet exist? I'd suggest not. The benefit can't be measured in such a way. The LNP are luddites not because they are lazy but because their thinking is decidedly 1950's. The NBN replaces a system 100 years old in parts but the LNP say if it isn't broke don't fix it. The truth is copper, telephone exchanges and analogue signalling are technologies way past their used-by date. The places in the world who have FTH are thankful for it. Why? because it will be as fast as developing technologies allow whereas copper has reached the limit. Wireless, as we know, is subject to atmospheric conditions and towers quickly become overloaded. Imagine how annoying dropouts would be if you depended on wireless in any form as your sole net access point. The NBN was always a good idea and the LNP opposed it only because they oppose everything like a child fighting for a toy. Just like the Carbon Tax which was Abbott's own idea, they want it until they have it, then they don't want it any more.
        • "Had a CBA been done on the Internet itself, would the Internet exist? I'd suggest not."

          Perhaps the best comment in relation to CBA's I have read, kudos, gaycarboys.

          This basically brings everyone of the ant-NBNers silly arguments crashing down.
  • Like most things in life, the devil is in the details. If a cost benefit analysis included a societal element, I'm certain nobody on either side of politics would have an issue with it.
    To look at it in purely short terms economic terms would be the equivalent of arguing that there isn't any point in creating a national telephone network, as the telegraph will work just fine for the next three years (read: election cycle). It's this kind of idiotic, myopic thinking that the NBN needs to overcome.

    Tenuous analogies aside, it's encouraging to see the way the Coalition (or Malcolm Turnbull at least) is repositioning himself/itself slowly on the matter. It would be smart for the Coalition to support the NBN in its current form- it's popular with the public and by throwing its support behind the FTTH rollout, it's one less reason to vote Labour. I think they can see the writing on the wall and are ramping down the attacks on the NBN so they aren't seen as backflipping come election time. One can hope in any case, as it's too important a piece of infrastructure to be used as a political pawn.

    I guess we'll only know how this is going to play out a little closer to election time, but if it does play out as described above, it'll be interesting to see how the anti-NBN mouthbreathers on this forum spin the policy change, having been so anti-progress for the last 18 or so months.
    • Good points; but how do you establish consensus about the terms of reference of a cost-benefit analysis? What is to be included? How far in the future do you look for benefits (or costs)? What costing do you apply to "social" benefits? How do you account for transformative technologies and industries that would be directly enabled by the NBN, but are not quantifiable today?

      A CBA works quite well in some situations - for example, when a civic planner confronts the everyday problem of how to get traffic across a river: do you (a) build a bridge, (b) dig a tunnel, or (c) let drivers keep using the old car ferry because "it's cheaper"? The terms of the problem are quantifiable, the costs can reasonably be estimated, and the benefits of the project can be defined in quite precise terms. That's where a CBA works well.

      Where a CBA does NOT work well is something on the scale of the NBN - groundbreaking and sweeping infrastructure development PLUS sweeping industry reform, all at once. Governments are often accused of thinking small, of only thinking about the next election, the popular benefits, the quick fixes. The NBN is the very opposite of all these things. It is a set of reforms and an infrastructure build which rest only within the capacity of government.
  • I think it's for the very reasons you mention in your first paragraph that there is no CBA. With the ideological differences and vested interests involved it would be impossible to get a consensus on the terms of reference (let alone your other points). I probably should have been more succinct- what I was really trying to convey was that if a CBA reflected the net benefit to society over it's operational lifespan, everybody would be happy. The Coalition would get their CBA, and Labour would be vindicated as it would show an enormous (if at all quantifiable) benefit.

    Of course, it's quite likely that the Coalition know this- if so, the CBA angle isn't anything more than cheap politicking. This seems more likely as the scope and projected cost of the rollout has remained constant from day dot, yet the Coalition seemed to have dropped the CBA angle completely,
  • I think the CBA point here is fairly much moot now. There was some, limited, argument for it before the NBN began, but as many people have stated it would not have been a fair comparison as it does not take into account social benefits.

    I think the Coaliton have realised this and that's why they've stopped voicing the point. Problem is, politicians going quiet on something, doesn't mean when they get the power they won't turn around and decide to do it anyway....
  • You would think so, but after this post went live Turnbull's office finally got back to me and said that, if they win office next year, the CBA is still on like Kong.

    They will not, however, be talking about terms of reference until some time before the election. Next week is "before the election" but I think they mean late next year. Let's just hope they give us more than 11 days to digest it like they did last time!

    Of course, that also means we have well over a year to speculate.

    Their determination to press on with a CBA is interesting on many levels, not the least of which is that they've all but conceded they can't stop the rollout. How does one construct a CBA for a project that's already underway? Anybody here who has done this kind modelling?

  • I think that a CBA is unlikely because with the high proportion of customers now electing for the highest rate (50% of connections in April) any CBA with reasonable frames of reference is going to come out strongly in favour of the current NBN model.

    Less than 18% of NBN connections have been at ADSL speeds. That leaves a whoping 82% of customers that cannot be serviced by the coalition alternative. Fully half could be serviced by no other technology than fiber.
  • In a way, it's comforting to see how far Turnbull has moved in the last 12 months. If the government goes full term, that gives us about another 18 months in which the Opposition's NBN position will undoubtedly shift a fair bit more as more and more of the electorate start to realise the enormous value of a high speed ubiquitous telecommunications network.
  • If Messrs Abbott and Turnbull really have an alternative universal broadband solution, why don't they commission a cost-benefit analysis of their proposal? Of course, their plan has to address not only HFC and ADSL in cities but also backhaul, satellite and wireless costs, state what bandwidth will be guaranteed to premises in regional and urban fringe areas, and give indicative costs for households and businesses to connect at various speeds and data volumes, as NBNCo published in May 2010.

    Now, here comes the kicker. The cost will be higher, not lower, and the service will be inferior.

    On construction cost alone, and even without factoring in the costs of ending the NBN, it will be more expensive than the NBN to deliver even 6 Mbps guaranteed to all regional premises, let alone the 12-25 Mbps they will get from LTE wireless and the NBN satellite. For this reason, I expect the coalition to seek a face-saving way not to terminate the Ericsson LTE contract nor the satellite construction. And within the 93% urban fibre footprint, the miserable upload speeds of laternatives to fibre will be such a burden to the economy that it will be impossible to justify the building and maintenance of copper and coax. Finally, electricity costs per gigabyte delivered over FTTP networks are 80% less than ADSL/FTTN.

    If the coalition's alternative - and it must be the universal alternative, as they are very sneaky about talking beyond one technology at a time - is tested with the CBA standard they insist must always be applied, then it will be shown to cost more, and remember that it will be an on-budget expenditure.

    Even without considering its inferior outcome, the higher cost alone should rule out making any change of direction. But of course we are taliking about partisan politics, not good governance. As a coalition sympathiser, I am disgusted by the waste of parliamentary resources and the eight month delaying tactics of critical infrastucture from Abbott and Turnbull and would not wish to entrust prime ministership to either of them. They are counting on the alternative being even more distasteful to the electorate, but they expected that to get them over the line in 2010, too.