NBN MTM CBA: Three Letter Acronyms For A Political Football

The long-awaited NBN Cost Benefit Analysis was released overnight, and to nobody's surprise it paints the multi-technology model (MTM) favoured by Malcolm Turnbull in the most positive way possible. The problem is that the CBA also appears to make all kinds of crazy assumptions in order to score political goals.

Image from: Dion Gillard

The NBN is a political football.

It's been that way ever since it was mooted, having been kicked around between the two major political parties, with a few stray fouls from newspapers and talkback radio along the way. For a football that's going to cost billions, you'd think we were getting something shiny and high-tech, but instead what we're left with is increasingly looking like a tired lump of over-inflated pig's bladder with worn out leather creaking at the seams.

Which is the colourful way of saying that there's no way of analysing anything to do with the NBN and its emerging state without a form of political debate and mudslinging along the way. The sad thing is that for a policy that could propel Australia's economy forward in all sorts of interesting ways, nobody comes out looking good from either side. The Vertigan Cost Benefit Analysis is just the latest in a long line of attacks that mixes a variety of critical details with a number of assumptions that heavily favour particular use cases and scenarios.

You know, politics. It's a grubby game played by both sides of the equation, and it would be ludicrous to paint either side as being virtuous angels at any degree. Pretending that the FTTP rollout under the previous Labor government was all sunshine would be to ignore the very real facts surrounding missed deadlines and fudged figures.

In many ways, the Vertigan CBA is just another in a long line of fudged figures, although a lot of those figures not only work to retroactively justify the MTM model, but also make some fascinating insights into what the committee felt that broadband was for.

I'm annoyed by all of this, if it wasn't already obvious. I do understand that a CBA is designed to look at net returns based on a costs model, and that can have some utility. Equally, there is obviously only so much money in the pot for a government to spend, and it is in the best interests of everyone that they're spent as sensibly as possible. Depending on where you choose to cherry pick, it's feasible to dig out figures on government waste from any administration. Don't like the Liberals? Talk about expenditure on defence budgets based on possibly wonky planes. Don't like Labor? Talk about insulation schemes.

Having said that, the overwhelming theme of the CBA is the idea that a government project should bring a net "return", as though government was a pure business, and not instead a governing body for the good of the people. That irritates me no end, because the role of a government isn't to provide some net "profit" for the people; it's to provide a common good. Under the "profit" logic, we should probably dismantle large swathes of Medicare and let people die, to pull that logic to an extreme end.

There are other assumptions and fudges that make me angry too. While much of the report focuses on download speeds and expectations, there's significantly less focus on upload speeds, on the general assumption that the needs for uploads are lower. That may be true — we probably don't need absolutely synchronous downloads and uploads in any case — but to suggest that there won't be an upwards curve for uploads in both consumer and business cases is quite laughable, not to mention dangerous in the context of an increasingly digital world.

Then again, we're not actually allowed to know what the real expectations of FTTN uploads are anyway. There's a table on page 46 of the 196 page report that states the expected download and upload speeds for each technology type, which puts FTTP in the over 100Mbps category for downloads and over 50Mbps category for uploads. FTTN is stated as "Mainly 50-100, some 25-50 and lower" for downloads and "Mainly 20-50, some lower" for uploads, which again would seem to match with what we know about likely FTTN scenarios.

Except that in preparing the report, the panel actually had access to more data than that, noting " the panel had access to FTTN download and upload speeds by distance from the premises to the node, but these data are not included here for commercial‐in‐confidence reasons."

Would it be cynical of me to presume that might be because FTTN drops speed calamitously the further you get from the node, and that might not sit well with the job of justifying the MTM model? Then again, that relates back to the expectations of actual data usage, which again are on the very low side and presume that customers aren't interested in or willing to pay for higher speed tiers in any case.

I'm not convinced that's true in any real way. Early NBN models predicted much the same thing, in that those connected would predominantly take up the lower speed, lower revenue 12Mbps plans. Even by 2012, though, 44 per cent of those on an actual NBN plan had opted for the 100Mbps tier. That's actual data from actual customers, not predictions or assumptions, by the way.

It's fascinating, given the CBA's particular modelling, that it's opted to use the $35.3 billion pricing model for an FTTP rollout, given the near assurances that the Coalition had around FTTP costs being around the $90 billion mark. Instead, the CBA has essentially the figure the previous government used. As such, the gap between the two, ignoring any productivity benefits of an FTTP rollout which the CBA assures us just aren't there is some $16.1 billion. That's the net return difference that the CBA assumes.

But there's another assumption that bothers me there. The MTM model is presented as ideal due to the faster rollout speed and lower cost, but also due to the fact that it can be upgraded at a later date. The issue there is that there's very few real details provided for what that likely cost is going to be to manage that upgrade to an FTTP path. There's mention of a cost saving, in that they assume that 20 per cent of the cost of upgrading to FTTP would be eliminated by rolling out FTTN first, as per page 88 of the report, but unless I'm missing something obvious, there's no actual dollar figure attached to that likely upgrade cost in the future.

The reality here is that while the actual core technology and fibre may indeed become cheaper over time as technology progresses, the cost of the labour to do so isn't likely to drop in price at anywhere near the same rate. As such, by failing to take into consideration any kind of likely upgrade path costing the CBA ignores a genuinely critical component of any kind of reasoned analysis. Given the scope of a real upgrade, it's not impossible to suggest that the gap between the two could eat up that costing difference quite easily. It's apparently perfectly fine to label the FTTP solution as near cost-neutral to doing nothing at all, but not to assess the costs of redoing labour all over again as part of an upgrade path. That's what a truly politically neutral CBA should have done, but then I'm probably being optimistic there in assuming such a thing would ever happen.

Some NBN opponents bleat at this point about the benefits of wireless as some kind of universal panacea — in fact, Kim Williams, ex-CEO of Foxtel/News Corp Australia makes that entirely stupid point in The Age today — but that ignores both the physical limitations of wireless technologies and the fact that you need to have some kind of network behind the wireless to pump data along the airwaves. I'm certain that particular chestnut will continue to be kicked around, as will the CBA in the coming days from both sides.

As I stated at the outset, it's impossible to hold an NBN discussion without it becoming a massive political scrum. For what it's worth I don't hold that the MTM model is entirely worthless. There's clearly some benefits in scheduling rollouts to areas where there's greater need and lesser access as part of a wider scheme. But the CBA sits in a position where it's essentially justifying MTM as a means to an end, not a road along a technology path.

Meanwhile, Australia sits with broadband that ranks 56th in the world for download speeds and 96th in the world for upload speeds according to Ookla's Net Index. While the NBN political football gets kicked around, we're at best standing still, and realistically falling behind with no clearly defined goal in sight.


Comments

    I'm done waiting years for the NBN.

    I'm getting fixed wireless broadband installed at my place next week, and (all going well), I should have faster than the 4-8mbps speeds I receive from ADSL now.

    I live in a low income area of Adelaide's northern suburbs. Telstra etc will never roll out FTTN or HFC to my area unless the government force them to do it. Ironically, mobile phone networks are frequently heavily congested in the area, and many people have that as their only source of internet.

    Despite having almost a dozen towers in the area already, Telstra have submitted plans to council to build 2 more 4G towers, at Davoren Park (to cover Davoren Park and Elizabeth North) and Elizabeth Park (to cover Elizabeth East).

    Anyway if i'm wrong and FTTN does finally make it to my place sometime soon, well i guess i'll be looking at spending another $300 or so in setup fees to get it connected.

    The upload rate of FTTN is what most concerns me.

    Earth to Malcolm: Some of us would actually like to use cloud services for backup!

    Last edited 27/08/14 10:57 am

      My area, new greenfields are getting NBN to the house and everyone else in older houses will get Wireless Internet...This is stupid to me but that's life when Gov is involved

      I'm actually jealous of you. I live in an area where adsl2 is not available, nor is fibre or fixed wireless. Just crappy 1.5mbps adsl1 and extremely patchy mobile coverage.

      What's even more insulting is that I have a clear line of site across a valley to an area where I know there is fibre. Yet fixed wireless insists on using LTE for the backhaul, at least according to the very vague documentation on the subject.

        There are already several alternatives to NBN fixed wireless in each state. Look into what is available in your area:

        I live in SA so I know for certain here that these providers provide fixed wireless (NOT NBN)
        Adelaide: Nuskope/Adam WiMax - most of metro Adelaide is covered.
        Regional SA: Nuskope/Adam/Internode (also known as Agile Regional Broadband) - all in different towns, depends on where you live.

        By searching for providers in other states I was able to find:
        NSW: W3 networks, SkyMesh
        QLD: Skymesh (everyone else seems to be 3G/4G MBB and NBN Fixed Wireless)
        WA: Ocean Broadband, SkyMesh

        A 10Gb 2-way satelite connection goes for $120 with speeds up to 6 megabit on SkyMesh.

        Depends on what your needs are, and how much you are willing to pay.

        Have you looked into "Reach" ADSL2+ plans? Or ADSL2+ direct from Telstra? With the newer standards, in theory anyone who could get ADSL1 in the past should now be able to get ADSL2+ (although the price is obviously higher)

        Last edited 27/08/14 12:28 pm

        I'm jealous of you! I can only get mobile internet out here, and christ all mighty it's expensive. I'll take the cheap slower broadband with big bandwidth to the shoddy data allowances if fast speeds of mobile broadband.

      But you only need 2 Mbps for real time cloud backup, even if you're a business user. The cost-benefit analysis says so!

        Of course, for all your little word documents and excel spreadsheets.

        Lets ignore anything created with Adobe or video. It took me a good bit of time to upload our Lync presentation to Sharepoint, it was only 60mb!

          An SSH session to my universities computer science servers has a noticable lag when typing. Woooo.

    Its okay tony! Im happy on 200kbs for gaming and youtube. No rush.

      Gaming is seen more as a luxury, so to complain about it is just #firstworldproblems - not that I don't agree with you (bf4 on adsl1 sucks).

      The argument needs to be "Here are all the benefits to all australians. Gaming without lag is just an added bonus"

    Great points! The upgrade cost and underwhelming performance are really concerning. Not to mention the underlying assumption that while either of the big models will pay for themselves, the smaller one does so faster and therefore should be lumped onto people regardless of any other considerations...

    This a political CBA to prove a political agenda not a proper independent audited CBA.

    3 things:

    1. Cost to upgrade from fttn to fttp in the future ommited from CBA
    2. Cost to maintain copper ommited from CBA
    3. Cost to replace bad copper

      I understand that the benefits and costs were also narrowly defined (in other words lots of things excluded), and sensitivity analysis on potential industries etc. as well as cost associated with new businesses (such as those reliant on such speeds) no longer being viable is not incorporated.

      Maintenance costs of the copper network is somewhat understated.

      It's a highlight of how CBA is often just used to justify a decision now rather than to provide a tool to an inform a decision.

    Despite it all, I'm less than a kilometre up the road from the last house with FTTP in my area, and because the copper will terminate to Fibre, I will be seeing some tiny benefits.

    But come on, I will be paying for the highest teir NBN plan I can get, I want to pay you money for this service. Tax payers are paying for the construction of this service. We are paying for it and we want it, why can't we?

    I agree with the op's comments in relation to Medicare
    We should let some people die
    There's to many hangers on ;)

    It seems to me 'internet' was able to progress so well and fluently back in the days when we were struggle to even download 10mb and like I said we progressed fluently and pretty quickly to where we are now. But as soon as politics came into play with how the internet is "handled" we have seen the shittiest progression within one of the best countries that has luckily been kicking hard regardless of how America treated world economics.

    And because of politics the nbn has been the biggiest bullshit since myki was being implemented....come on not even one state government could properly switch from paper tickets to plastic tickets with out it costing billions over the budget and constantly being screwed for various reasons and with saying that we've trusted a government to run the upgrade of the whole nations internet?

    Does any one not realise we are the ones who should dictate what the government 'does' and not the other way around? We as citizens are still the majority as the government is still and will always be the minority.... I think as this country and a lot of other countries have forgotten that.

    Would it be cynical of me to presume that might be because FTTN drops speed calamitously the further you get from the node, and that might not sit well with the job of justifying the MTM model? Then again, that relates back to the expectations of actual data usage, which again are on the very low side and presume that customers aren’t interested in or willing to pay for higher speed tiers in any case.

    Nope ... not cynical at all. The drop off in speed the further you get away from the node with be on par with the drop off in speed that we currently get the further we are away from an exchange. It's just that the nodes will (hopefully) be closer.

    20 per cent of the cost of upgrading to FTTP would be eliminated by rolling out FTTN first

    I read that as implying the upgrade cost to FTTP would be 80% of $35.3B, i.e. $28.2B.

    Now add that to the $29.2B FTTN cost, for a total of $57.4B after the inevitable upgrade.

    So we're deferring a cost of $16.1B for the next 10 years or so, at the cost of paying that plus an extra $22.1B in 10-20 years. Also, there's the lost opportunities cost for being behind the global bandwidth curve vs. having $16B now (which was always going to be serviced debt anyway, not coming out of our cash reserves).

    I used to work for the state government and had to provide data for the premier at one point. The report that came out of it didn't actually resemble what the raw data showed. The raw data showed that a plan implemented by the previous govt had proved successful and was steadily improving over time. Instead they cherry picked only facts that they could use in isolation to back their claims. While the information they presented was technically true it wasn't the whole story and painted the previous govt in a negative light and them in a much better one.

    *sigh* Sadly that's politics for you, it happened then, it's happening now in this report... just pick the facts that look palatable and fail to report on ones that don't.

    Your complaint about measuring "returns", which you then conflate with "profit" is off base, as shown by your Medicare example.

    When looking at a huge infrastructure project like this, of course you should look at whether it provides a net return - otherwise the government shouldn't do it. People like to compare the NBN with the power grid, roads etc, and in that respect the logic is just applicable. If deciding whether to spend billions on a road (for example), then you look at whether it provides a return which justifies the cost. You can't just say "its about providing for the common good" - you need some way to quantify whether the common good is being sufficiently improved by the proposed expenditure.

    Return is not the same thing as profit. A public road will never turn a profit, but a study on whether it should be built will still look at whether it provides a net return.

    "Another in a long line of fudged figures"
    Yes - let's compare them to those in the cost-benefit analysis that Labor did before they started on the whole scheme.

    Oh, wait.

      Irony: Labor's CBA process was just as useful, and a lot cheaper.

    I have a something I want anyone to answer me;

    If the Liberals are now using MTM and that 30% of the NBN will be using the existing PayTV cabling, why is it a cost benifit that we spend $47billion for only 70% of the same coverage as the FTTP rollout?

    Seems to me that the FTTN is more costly than FTTP when 93% coverage is included in the total costs.......

Join the discussion!

Trending Stories Right Now