“Half true”: Politifact partially backs Turnbull’s $94bn NBN figure

103

turnbull

news Fact-checking media outlet Politifact has rated the controversial claim by the Coalition that Labor’s National Broadband Network could cost as much as $94 billion as “half-true”, as debate continues to swirl around the veracity of the Coalition’s own figures.

The Coalition has been making the $94 billion claim for around the past six months and especially since the launch of its own National Broadband Network policy in April, arguing that the fundamental assumptions underpinning Labor’s NBN costings are inaccurate and that the network is likely to cost much more to build than the $44.1 billion Labor estimates.

According to the Coalition’s background briefing paper (PDF), the $94 billion number will come about as the result of a number of Labor assumptions being wrong. The Coalition states that for the $94 billion figure to eventuate, NBN Co’s revenue must grow much slower than currently forecast, construction costs must be significantly higher than currently forecast, more households must pick wireless alternatives than is currently forecast, and the NBN must take 50 per cent longer to build (an extra five years) than currently forecast.

Commentators such as the writer of this article have argued that these assumptions represent a “worst case scenario” for the NBN where every cost factor blows out, and that as such, the Coalition’s premise is not reasonable. And senior Labor figures such as then-Communications Minister Stephen Conroy have accused the Coalition of lying” about the cost of Labor’s NBN and “concocting” financial figures.

However, in April Shadow Communications Minister Malcolm Turnbull responded to this claim, stating that the $94 billion figure being cited is not a ‘worst case scenario’, with the Coalition estimating that Labor’s NBN could actually cost more than $100 billion and take 20 years to deliver.

“The NBN has missed every rollout target it has set for itself. So is it so unrealistic to assume the rollout won’t be complete until 2025, four years late?” the Liberal MP asked at the time. “We also modelled NBN revenues. NBN Co assumes it can increase broadband access revenues by 9.2 per cent a year in real terms, and in doing so increase the share of household income devoted to fixed-line telecoms by 60-70 per cent. Our more realistic assumption (shared by independent telecoms analysts) is that its share of the wallet will stay constant.”

“We modelled the percentage of wireless-only households rising to 25 per cent rather than the 16 per cent assumed by NBN Co – reflecting the fact last year the number of wireless broadband accounts in Australia for the first time exceeded the number of fixed broadband accounts. And we modelled that the cost of passing each house with fibre would be 40 per cent higher than the NBN had estimated, to bring it more in line with independent analysts’ forecasts and the on-the-ground experience of Telstra’s fibre rollout in South Brisbane.”

“Our assumptions that led to the $94bn cost are far from a “worst-case” scenario; in reality the final cost could easily exceed $100bn and the rollout take 20 years to complete.”

A new analysis published by Politifact late last week (we recommend you click here for the full article, it makes for very interesting reading) concluded that Turnbull was “on strong ground” when he said the final cost of Labor’s NBN would likely exceed $44.1 billion, as most major projects run over budget. However, Politifact added, it was difficult to estimate the total cost of Labor’s NBN project.

Politifact said in general that Turnbull had “overreached somewhat. “If he had merely stated it was highly probable Labor’s NBN would end up costing more than the claimed $44.1 billion then, based on what is known he would be hard to fault,” the site wrote. “Turnbull might be taking things out of context, but based on our research and analysis, there’s more than enough of a legitimate challenge to NBN Co’s corporate plan in his estimate to push it out of False territory.”

The news comes as debate continues to rage between the two major sides of politics about the extent to which each side has properly verified the financial details of their NBN plans. For example, this week and last week Turnbull has avoided directly answering Labor’s question of whether the Coalition will submit its alternative National Broadband Network policy to the Treasury or any other organisation for costing purposes, instead accusing the Labor Government of not being transparent about its own numbers.

Turnbull believes that the Government is unfairly withholding NBN Co’s latest Corporate Plan from being published until after the election, although Communications Minister Anthony Albanese has denied that NBN Co has delivered the final copy of the document to the Government.

opinion/analysis
My view on this issue is well-known, and to be honest, it’s pretty much exactly the same as Politifact. I think it’s clear that there will be some variability around NBN Co’s budget, and I think anyone who doesn’t expect to see at least a 5-10 percent variance, for an infrastructure project of this size (the NBN is unprecedented in both size and complexity in terms of Australian infrastructure projects) is fooling themselves. I’ve been reporting on major government technology projects for most of the past decade. I have yet to see one which runs perfectly to time and budget.

However, I also believe Turnbull’s $94 billion claim to be unsubstantiated. As I wrote in an article for the ABC when the Coalition released its NBN policy:

“Among politicians, Turnbull’s speeches and policy workings are unusual in that they are usually exceptionally – if sometimes conveniently – well-referenced. However, the background briefing document released this week does not boast a sufficient bibliography to back its claims. A good example is its argument on the cost of connecting fibre to premises, which it baldly claims is $4,000, based on a single report published by Macquarie Bank’s equities department in January.

Conroy claimed this week that the Coalition had “concocted” its NBN figures. I don’t think we should go that far. But it also true that the Coalition’s claim that Labor’s NBN will cost $94 billion does not yet stand up to scrutiny, based on the available evidence.”

In short, we should expect NBN Co’s costs to blow out. But there is very little evidence that they will blow out to the extent Turnbull and the Coalition are claiming. I support Politifact’s conclusion of “half true” in terms of Turnbull’s claim, meaning “The statement is partially accurate but leaves out important details or takes things out of context.” Another fair rating would also be “mostly false”: “The statement contains an element of truth but ignores critical facts that would give a different impression.”

It’s certainly not in “pants on fire” territory.

Image credit: Office of Malcolm Turnbull

103 COMMENTS

  1. From what I have seen of politifacts they are slightly right leaning and weight they ratings accordingly on that basis if you take their rating and move it one down for Liberals and one up for Labor is is actually quite accurate.

    • +1

      This has been my observation as well. A fair amount of their analysis is balanced, but it does slightly lean to the right at times. That being said, i think they make some valid points with the NBN as Renai points out.

      • I would agree except where it comes to the NBN.
        As far as the NBN is concerned their ratings seem to be completely inconsistent with the reasoning laid out in the article accompanying it.

        • Their reasoning is spot on and well thought out, you may not like it because you want the Labor NBN at any cost or any delay, but that’s not a reason to dismiss the Politifact ruling.

          • Really even Renai in the article says he would say mostly false and previous articles had flat out called the figure a LIE.

            Could the project go over budget Yes
            Could the project go over by that much based on the evidence No

            The question is will the NBN cost $94 Billion or more? the answer is No

            This is the Fact this is the answer to the question whether Liberal fanbois like it or not

          • OK WTF is going on it seems the original Politifacts article published in The Age
            Is Substantially different in its conclusions

            Original

            PolitiFact finds Turnbull’s claim that Labor’s NBN would eventually cost $94 billion possible but unverifiable. It rates it “half true”.

            Read more: http://www.theage.com.au/federal-politics/fact-checker/will-the-nbn-cost-94-billion-20130816-2s154.html#ixzz2cYTbEn7Z

            Politifacts

            Turnbull might be taking things out of context, but based on our research and analysis, there’s more than enough of a legitimate challenge to NBN Co’s corporate plan in his estimate to push it out of False territory.

          • Ahh the irony when you are calling him biased for reporting on a quantitative statistic.

            It does show a logical fallacy and another measure of bias.

          • “Ahh the irony when you are calling him biased for reporting on a quantitative statistic”

            Michael

            And the irony of you calling a few arbitrarily selected finding, quantitative statistic.

          • Don’t worry,

            I’m sure you will understand how statistics work one day.

            Btw how is an entire population a “few”?

          • Michael

            I don’t think your current and previous demonstrations of your understanding of statistics and research methodology warrant you being patronising.

            “Btw how is an entire population a “few”?”

            A few findings do not represent an entire population.

          • Thanks for highlighting my point.

            A population is not defined by the number of observations within a set but by the proportion of those observations used.

          • “A population is not defined by the number of observations within a set but by the proportion of those observations used.”

            It is always nice to be lectured by someone with dictionary understanding of research.

            The part you forgot in your simplistic definition is that these so called observations are meaningful when randomly selected.

            These are not random examples of false findings. Furthermore, the scientific value of these findings is highly disputable for reasons I have stated earlier which you are welcome to refute but haven’t as yet.

          • You are confusing different issues.

            Randomness, is relevant to samples. When taking a sample from a population you need to ensure that it is not biased so the simplest way is to take a random sample. We are dealing with an entire population so that problem is avoided.

            The issues surrounding politifact I agree with, but they affect the information gathered from the data not how you analyze it.

          • Randomness, is relevant to samples.

            If your saying Fibroid (the post that started this) is a “random sample”, then I fully agree :o)

          • “You are confusing different issues.”
            No I am not

            “Randomness, is relevant to samples. When taking a sample from a population you need to ensure that it is not biased so the simplest way is to take a random sample. We are dealing with an entire population so that problem is avoided.”

            How could a few self selected topics represent an entire population of “false statements”? An entire population would comprise an analysis of all statements made by politicians, not just a few selected one.

            Are you suggesting that because Fibroid counted a few more Labor statement this makes it statistically significant?

            You started this argument by trying to use your strange interpretation of statistics to support Fibroid attempt to suggest that Labor lies more than the coalition.

            “The issues surrounding politifact I agree with, but they affect the information gathered from the data not how you analyze it.”

            So, let me get this right and try to follow your logic. The findings are flawed but once you analyse them it does not matter any longer? In other words, the information which was gathered cannot be relied upon but you can still conclude that “Labor politicians are more likely to lie”

          • “So, let me get this right and try to follow your logic. The findings are flawed but once you analyse them it does not matter any longer? In other words, the information which was gathered cannot be relied upon but you can still conclude that “Labor politicians are more likely to lie”

            What a tangent. All I said was that irrespective of your data, the same methodologies is used in analysing it. Can you please point out where I mentioned what you do with it once you have analysed it? I want to see how you formed that conclusion.

            “How could a few self selected topics represent an entire population of “false statements”? An entire population would comprise an analysis of all statements made by politicians, not just a few selected one.”

            A population is any defined set that share a common trait from which data can be drawn. They can be large or small. Some include more information and some less. Can you find a definition to support your view that they are only large? A population can easily be part of a larger one, (of which almost all of them will be.)

            The population considered:

            “It’s simple, I picked that ‘one’ because Labor represents the majority of false rulings.

            http://www.politifact.com.au/truth-o-meter/rulings/false/

            So the population is the Politifact “false” rulings.

            From there is it quite simple to look at them.

            I have no idea how you even considered that it was about ALL political statements (I
            ‘m assuming its in Australia) as no one has even tried to count that, or even sample it. Let alone judge them.

            But that would be another population of which this one is a subset.

            Then you can consider all political statements made worldwide as a larger population again.

            This is what there was data on, what was being used and so it was looked at.

            “You started this argument by trying to use your strange interpretation of statistics to support Fibroid attempt to suggest that Labor lies more than the coalition.”

            Where did I say that?

            I said that they had more results on the page in Politifact. How that translates to the real world is up to you. From your reaction I can see it has struck a bone, but as you have obviously missed it; It does not represent the larger population, nor does it even cover all that Politifact has looked at so I draw no conclusions beyond what I actually said.

            Labour has more False rulings that the LNP.

          • I am preferring the ABC’s fact checker site as they have better classifications for their facts.

            Politifact are changing the context of what Turnbull said for some reason and then rating it as Half True. Turnbull was not arguing that the cost may blow out by some amount. This would most likely be true. He specified 94 Billion and stated this was not the worst case but the probable case…. If he stated the blow outs could be up to 94 Billion or higher, then maybe a half true. What he stated was mostly false at best.

          • Do you disagree with all of Politifacts rulings or just those in your view that are not critical of the Coalition enough?

          • *shrug* No one is critical of the Coalition enough. You, yourself, should be more critical of the Coalition. That’s the duty of a responsible voter.

          • I hate fucking elections.

            One turns on the TV and is bombarded with political crap.

            Then the radio and more political crap.

            I popped into the post office and in the car park was a local pollie and the team, there canvassing support.

            Everywhere one turns, election.

            Ah thank god for Renai and all at Delimiter….

            Well… apart from one :(

          • There are many politifact rulings that are not preferable to the way I will vote at the upcoming election, and there are many polifact rulings that are preferable to the way I will vote at the upcoming election.

            In either case, the majority facts analysed point to a sad state of affairs for Australian Politics, though watching Jon Stewart, I must say I still prefer our system to that of America. (Google – Daily Show Monsanto)

            There are some dubious rulings though where they seem to be a category or so off the mark, and for casual observers this can make all the difference as they are likely to take rulings at face value.

            – Gold Plated Coffee Machines was a joke, no-one expects it to literally mean coffee machines actually being gold plated, and that was not the point the comment was trying to make.
            – GetUps ruling of Mostly False for Australias rating of 62nd in Refugee intake per capita is not inline with other rulings they gave. – How can it be mostly false when the figures check out, but it was only telling part of the story? That is clearly half true by tPolitifacts own specified standards.
            – I could argue the pants on fire rating against the CDP….. but its the CDP and at best should have been given a False anyway.

            The two that really stand out to me, given the stark difference to the way they were handled are:
            – Rudd stating that 70% of boat arrivals were resettled in Australia. This is a plain example of a specific
            figure given and a rating of False because in reality although 70% were resettled, only 43% were in
            Australia. Again, it was false because of the number specified was incorrect, despite there being more to
            the story than this.
            – Turnbulls Half True for NBN blow out to $94 Billion dollars is similar in reach to that of Rudd’s statement,
            yet Politifact shift the posts to “Turnbull is on strong ground when he says the final cost of Labor’s NBN
            would be likely to exceed the claimed $44.1 billion.” In essence this is not the point being judges, as per
            the title of the article “Will Labor’s NBN cost $94 billion?”. Without being able to verify Turnbull’s
            numbers, and very little to discredit the current figures of $44.1 billion, Turnbull was given a ranking two
            rungs better than Rudd, for a similar level of over-reach in their statements.

            On the whole I strongly agree that Politifact seem to lean towards the right in their rankings and outcomes, whilst overall still generally fulfilling their purpose of holding politicians to account for what they say. And although wouldn’t go as far as renaming them Politiscat, I still think their ruling on the $94B NBN is full of S#&T.

          • My problem with politifact isn’t necesarily their “results”, but their selective judgement.

            Sometimes, they “fact check” based on precise literal interpretations of words. Sometimes they “fact check” on close to reality.

            according to politifact:
            “it will cost more than 44 billion therefore 94 billion is half true”
            “You can be charged 5000 dollars to connect, therefore it will cost 5000 dollars is mostly false”

            In my book; they are pretty close to the same level of lie.
            In a literal sense; they are both pants on fire. Sometimes politifact goes super literal, sometimes it doesn’t. I just wish they would pick one and stick to it.

          • But the cause of concern for me is when they choose to be selective in their judgements. This is what is creating the appearance of leaning to the right of politics.

            You are right that if they were consistent with their approach it would remove most of the problem.

          • Much like Penny Wongs “guess” that the LNP need $70b in cuts was rated “False” (even though it says: Its finance spokesman Andrew Robb said on ABC NewsRadio that he would “need to identify up to $70 billion over the next four years if we are to get this economy back in some sort of shape”.)

            While Malcolm’s “guess” that the NBN will cost $94b was rated “Half True”…

            The actual articles where they detail their thinking are pretty good, but the meter part of it is junk.

          • The problem with their assessment is that they treat “estimates”, “predictions”, “speculations” as “facts”. If you’re calling yourself Politifact, you should stick to the facts, unless, of course, you have a crystal ball.

            An example of this is the $70b black hole. It is more than likely to be less than that, but there is a possibility that there will be a black hole. So, the answer should be that it is likely (not true or false) that there will be a black hole.

  2. Sorry, this is not even half true. Worst case, for sure, there is a chance, but The Turnbull has taken a small fact and extrapolated it into a massive 250% overspend.

    “The NBN has missed every rollout target it has set for itself. So is it so unrealistic to assume the rollout won’t be complete until 2025, four years late?”

    Every missed target is a result of a single event – extended negotiations with Telstra, and the butterfly effect caused by that. Its not an event that, by itself, expands over the duration of the rollout causing more and more delays, but a single event that moves the goalposts by a similar timeframe.

    Figure that 9 month delay, and the 6 months its caused elsewhere, and the delays are in the region of days, not months, or years like The Turnbull would have you believe. Hardly a cause for concern.

    Its disingenious to keep coming back to a static delay and giving it a dynamic effect, and its rulings like this that seem to continually favor the Liberals that make me question Politifact’s neutrality that little more each day.

    I still wonder why these sorts of events are only ever going to effect the Labor plan, and not the Liberal one though. 25% growth in wireless only use would seem to effect the shorter lifespan FttN rollout far more significantly, by reducing the population able to be gouged to pay for it.

    • We’ve just seen a $5 billion increase in construction costs announced, although Quigley claims other savings will cover this.

      Of more concern is the fact that to meet the 7% return ARPU will need to rise. That means the NBN plans will be more expensive, meaning less people will connect, meaning ARPU will need to rise, meaning less people will connect, …. You get the picture.

      • Not going to bother arguing with you about ARPU because no matter how many times someone addresses it, you ignore it.

        As for the $5b, frankly its something we’ll only see resolved if Labor wins on Sep 7. Its a long term guess on both sides – one sensationalising it, the other talking it down. Where reality is, who knows, but I’m not keen on believing the Liberals just because the AFR have decided its an unrecoverable increase in costs.

        Read Renai’s opinion on the situation, I think its spot on. Its also a good guide to the whole debate – Lib’s oversell the smallest issue, Labor does the opposite.

  3. If government projects have a tendency to overshoot their budget as you suggest, then presumably the same would apply to the Coalition version of the NBN.

    • I read a report into large government projects; that found on average; they overshot by approximately 10%.

      Some undershot; some spectacularly overshot, but I think it averaged out to 10%.

      Can’t for the life of me remember where I read it; might have been specific to QLD state government. At the very least it referenced the QLD state government for something.

  4. The problem I have around Turnbull’s figure is it looks like he has made the data look like it supports his position rather than the other way around. He has been saying for years that FTTN should be 1/3 to 1/4 of FTTP. Then he prices his plan, finds out it will actually cost around $30bn, then concocts a figure so Labor’s looks to be 3 times more expensive like he said it would.

    • +1

      There is no way Malcolm’s LBN figure is right, which is why he refuses to have Treasury take a look at it. I suspect the actual figure would be a lot closer (possibly even slightly more) to the current NBN price once he sorts out Telstra and node distances to reflect what he actually wants as a result (25/50Mbps “guaranteed” and vectoring/G.fast).

  5. If we assume the Labor NBN will cost twice as much as projected, then isn’t it safe to assume that the Liberal version will do the same? After all they still need to negotiate with Telstra, remediate the copper, build the cabinets, run the fibre, build the wireless network and so on. So will it cost 60bn and take an extra 5 years?

    • As far as I can tell, the best way to think about it is this:
      The liberals have committed to a price 29.5 billion,
      Labor have committed to a product FTTP for 93%.

      The liberals have admitted that the labor plan can go over budget, but categorically deny that their own plan can go over budget. The only logical conclusion, is that the liberals will only build as much of a network as can be afforded for 29.5 billion dollars.

      So given their predictions of cost overruns for the labor network will be 100% of the expected cost; it means only ~50% of Australians will get 25 megabits or more. (or whatever 29.5 billion dollars buys you for a 60 billion dollar FTTN plan).

      But hey; maybe when the liberals get in, they will fire all the contractors working for NBN co; and hire all the multitudes of skilled workers out of a job after NBN co lets go all its contractors to get them to build the new LBN, and since they will be fresh on the job they’ll do it faster and cheaper than the other guys!

      • I think you’ve hit the nail on the head.

        Labor’s approach was: Let’s see how much it costs to build a proper National Broadband Network, and then decide if we can afford it.

        The Liberal’s approach appears to be: We have already decided that we shouldn’t spend any more than $30 billion – what kind of network can we get for that?

        Thus, the circular arguments continue to circle. Labor points to the half-baked nature of the Coalition’s plan, and the Coalition squawks about costs and delays.

        I also find it deeply amusing that Turnbull has such great confidence that his own version of the NBN would not be subject to delays, cost blow-outs or contractor disputes. Perhaps he needs reminding that his plan also involves the installation of a lot of in-ground cabling near asbestos pits, using the same sort of tradesmen, and subject to Telstra’s interference and obfustication.

        • Also, Labor used an Expert Panel made up of experts from various fields to answer the question of what to do to address Australia’s future communications needs (and they came up with the FTTP NBN).

          It feels like Malcolm and his staffers came up with the plan to do an FTTN NBN to address the issue of the Liberal Parties future election needs, not those of Australians :/

          • The expert panel was only convened after Telstra refused to work with Labor on FTTN as an attempt to save face.

          • Incorrect…

            Telstra pulled out of FttN negotiations to build their own FttN network in August 2006…

            RFP’s to build the NBN followed (in 2008 iirc), where Telstra submitted a non-compliant bid.

            So Telstra actually refused to play ball with both sides of politics…!

            Perhaps they realised FttN, even as the incumbent with a last mile monopoly (being copper based) was dumb after all?

        • The problem with the level of politics of this country is that, being aimed at the less informed, it is discussed in terms of we are wonderful, the other party is awful.

      • @Peter you’ve hit the nail on the head.

        The Coalition won’t spend more than their $29 billion on the Turnbull Bullshit Network (TBN). When the money’s all spent, that’s it.

    • Stop putting things in context, dammit! We want fact-free outrage, thank you very much!

  6. Hmm half true…?

    Ok, I guess going by the meaning, a ‘statement meant to deceive”, we can live with that…

    Personally I’d call it bullshit, but then I’m not an Earl ;)

  7. The problem with fact checking political claims is that much of what is said by senior politicians is neither true nor false – it is so biased, misleading and devoid of context that it is, to quote physicist Wolgang Pauli, “not even wrong”. The question politicians ask themselves when deciding what to say is “Will this be convincing for my target audience ?” and whether it is true, false, or bears any particular relationship to reality at all is only considered in as much as being caught out lying looks bad.
    In this particular case, it is clear that Malcolm Turnbull has scoured the planet for figures that make FTTN look good and FTTH look bad. The whole exercise has been done in bad fath, and that is the core problem, rather than any particular number that he has produced.

    • “Politics is the art of the possible.” – Otto von Bismarck, 1867

      Politics is about making the impossible, or difficult, seem doable. Veracity doesn’t actually come into it.

      Malcolm is a very good politician.

  8. Nice to see this article is forward thinking…..

    …….although Shadow Communications Minister Anthony Albanese has denied that NBN Co has delivered the final copy of the document to the Government.

  9. There is no doubt that Labor’s NBN is going to cost a lot more than current estimates, it’s not possible to see into the future and guess what labour/fuel/equipment costs are going to be in 5 years and attempting to extrapolate today’s pricing into some period down the line. The same reasoning also applies to the coalition.

    The Abbott/Coalition “road” programs are all going to fail and will be cancelled in 2015, that’s when the demand for oil exceeds the supply of oil at current prices, when this situation occurs in the oil market it usually leads to very sudden and very large price increases which leaves the normalised pricing at the last historical high. Therefore in 2015 we will be facing a “normal oil price” price at $150 per barrel or $2 au per litre and spot shortages in some locations around the globe which could lead to spot prices spiking higher at times. The effects on economies will be more profound as airlines, plane makers and the automobile industry downsizes. All Government airport and road programs will collapse as the realization sets in that efforts to sustain the unsustainable with a business as usual approach is impossible.

    An advanced communications network is the only option we have to keep society functioning and it’s much better to start now rather than waiting till the shit hits the fan.

    I would expect the NBN as currently structured will probably cost around $60 Bd but could be made a little cheaper by making MDU’s pay for their own internal fibre (after all these installations are on private property) similarly private residences with excessive cable runs could be asked to contribute. The terminal box as Simon Hackett suggested could be simplified to a single Ethernet point.

    The total cost could also be kept down by The Government making a larger direct capital contribution initially to both speed construction (this also speeds revenue collection) and lessen borrowings to reduce interest payments.

    It’s my preference to see the government contribution for the NBN to be raised to $6-8 Bd per annum and the incoming government to delete the new maternity leave scheme (it’s 5Bd, the NBN is more important than middle class welfare), delete the last home seller’s grant which is sold as a “first home owners grant” saving 3 Bd, delete negative gearing on real-estate (it’s a rich dudes welfare system), delete all the other “small handouts” like back to school bonuses, delete the “baby bonus”, delete all aid and concessions (fuel) to the mining industry (another rich industry welfare system) private business should pay their own way.

    After all the middleclass, rich guys/industries welfare has been scrapped there is more than enough money for the budgets to be balanced and the NBN build to be built on an ongoing cash basis and the rollout to be sped up to around a 7-8 year time frame. Also it’s possible the NBN could be sold to TELSTRA and the NZ solution opted for with a division of Telstra into 2 companies one retail and one wholesale. With a direct government contribution subsidising the rollout of fibre.

    The Liberal Governments (as Malcolm Turnbull stated in his media appearance last night) view that the internet is an “entertainment system for movie downloads and piracy” is propelling their drive to do as little as possible for communications infrastructure.These guys still don’t get it and there obsessed with shovelling out welfare to buy middle class votes.

  10. I would like to see the tables turned around somewhat and that is if the NBN project is likely to blow out then why weren’t the various government committees which include Opposition members on to it. The answer is that they are all politicians and image (read spin and deceit) comes first.

    When I read of Politifacts opinion in this I wondered how they were so well equipped to counter the information that is being used to run and monitor the project. Is it more believable that a subsidiary organisation paying royalties to use a US brand name might be running an agenda or is it more believable that we have useless politicians of all stripes.

    Now I’m just confused.

    In my fantasy world if you want to hold government you would demonstrate your superior capabilities both in opposition as well as in Government. If you don’t / can’t then you don’t get my vote.

  11. I would be interesting to see how much the Coalition’s plan would cost if the same assumptions, Mt made, were applied to its plan.

    • How do you apply assumptions to costs when you don’t have anything but a total you pulled from your backside?

  12. I don’t necessarily disagree with any of this. But my only question would be- who decides whether they label it “half true” or “mostly false”?

    Because the actual wording of the rating conveys very different meanings. If someone said to me, it’s half true that a whale is a fish, I’d call bollocks on the rating. Because a whale is a mammal. It just happens to swim. Mostly false however, would be much closer. It does actually swim, like a fish, but isn’t classed as a fish. So you could say, I suppose, half true is literally 50/50 whereas mostly false would be 75/25.

    On that basis, I think Renai’s mostly false is more correct- the NBN is likely to blowout, or at least have to be pared back to keep it in budget if that’s required. But it is unlikely to cost anywhere even approaching $94 billion. Apart from anything, I believe the definition of Turnbull’s is hazy- he consistently says Labor’s NBN is going to “cost” more like $94 billion. Cost, in a business sense, colloquially means CAPEX- the “cost” to build the network. Turnbull’s own documents state the CAPEX in his equation for the NBN isn’t $94 billion- that’s the total funding. However, I will give it to him that he compares it to the total public funding requirement of the Coalition policy, rather than his CAPEX cost, which, while the $94 billion is simply bogus in my eyes, is the right figure to compare to his $29.5 billion.

    • Politifact are very inconsistent. They might decide to interpret a metaphor literally, such as ‘gold plated coffee machines’ and give a pants-on-fire rating. Or they might give free reign for a politician to use colourful exaggerations.
      They might demand that a politician get their numbers exactly right, or they might be satisfied that the general point was still valid.
      They can get pretty damning about trying to predict the future.

      Malcolm Turnbull has been given the kid-glove treatment on this one.

    • “half true” or “mostly false” is like being “partially pregnant”.

      While life its self has shades of grey, “True” and “False” are either/or, not degrees of being…the fact that they (Politifact) have degrees of truth show they are just as political as those the portend to judge.

        • I’m just referring to their “Truth-o-meter”, their “Our Ruling” section is usually pretty good I find there is generally a disconnect between the meter and the finding as True and False are not a “Shade of grey” concept (maybe it’s my programming background). “Mostly False”, is still “false”, as is “partially true”.

          The ABC fact checker unit uses a much better system for their meter IMHO

          http://www.abc.net.au/news/factcheck/

  13. I heard that God called Obama, Li Keqiang (Premier of China) and Abbott up to a conference last week.
    He told them that he was finally fed up with all the wickedness in the world and that he had scheduled the final judgment for Christmas day 2013.
    As the most influential Christian leaders in the world, they were to make this announcement exactly one week before the date.

    When they returned they each called their closest associates what had happened.

    Li Keqiang:
    “I have bad news and I have worse news. The bad news is we were wrong, there is a God. The worse news is that he will end the world on Christmas Day.”

    Obama:
    “I have good news and I have bad news. The good news is we were right, there is a God. The bad news is that he will end the world on Christmas Day.”

    Abbott:
    “I have good news and I have great news. The good news is that God himself thinks I am influential and we are going to win the election. The great news is, we wont have to admit that we were never really going to build an NBN.

  14. The more Politifact publish, the less relevant and useful I find them or their ratings system, for many of the same reasons as outlined by others. You can’t bemoan the inaccuracy of one statement based on an anally semantic interpretation, and then suggest a similarly inaccurate statement was ‘half true’ because some aspect of it has some liklihood of proving accurate (although even that is simple conjecture at this point). I don’t have any problem with the reality that it is difficult if not impossible to draw definite, unerring conclusions about statements of conjecture predicting the outcome of future events, but at least be consistent with your treatment of all subjects. Politifact? Seems more like politispin at this point…

  15. It is always amusing see that the accusation of right wing bias that is placed on any independent review that does not back the Labor rollout 100% or criticise the Coaltion plan.

    I am sure Politifact choose the wording of their final analysis very carefully to convey the meaning of the conclusion they make.

    Labor NBN supporters may want to choose different wording than half true because they are not comfortable with it to their own ‘feel good’ mostly false, if Politifact wanted to use mostly false they would have, as they have done in previous conclusions on different subject matter.

    What’s the point of independent analysis if you just re interpret it to suit your own point of view?

    Also half true does not mean you divide $94b by two to once again give you a result you feel comfortable with , it was not a numeracy based ruling.

    • “What’s the point of independent analysis if you just re interpret it to suit your own point of view?”

      Welcome to the Australian government, irrespective of party this is what happens. If they don’t like it they bury it. If they do they hold a press conference about how great they are.

      • “Welcome to the Australian government, irrespective of party this is what happens. If they don’t like it they bury it. If they do they hold a press conference about how great they are.”

        THIS

    • I accept the rating of half true. Half true in this instance means:

      With a different set of assumptions, the investment required will be $94b: True.
      The different set of assumptions is valid: False.

  16. The problems with Politifact are similar to those of another “independent” fact findings organisation, choice. They both cover many topics, which require specialised knowledge. Because of this, they often need to consult “experts” who have their own personal bias.

    On the question of the rating, truthfulness is like pregnancy. You can’t be half pregnant anymore than something can be half true. As well, what is the difference between half true and half false? It is a matter of interpretation.

    In this particular case, it is difficult to determine the veracity of the statement based on a possibility (not a fact). To say it is possible, or even likely, that the cost of the project will blow out does not make true, it only makes it possible or likely.

    The only way the statement could be true would be if the assumptions on which the costing is based are true then it is true that the project would cost 94b. Anything else is pure speculation not worthy of a true or false rating.

    • The only problem with Politifact is that they didn’t make a mostly false, false or pants on fire ruling in this case, if they did there would be minimal discussion to the length we have here as Labor NBN supporters would all be content that Politifact had in fact ‘got it right’.

      • The only problem with Politifact is they get it wrong by about 1 notch. Every time. And my perception is that its in the Liberals favor far more often than not. Want a simple example, take their review of the mining boom being over. Ruling is half true, when it should be mostly true.

        In the context of the statement, being about how we’re no longer able to rely on commodities and have to look at the next stage different.ly. Which is correct, we do. The article even agrees. Yet they get pedantic, and take it as a literal statement, and out of context, giving a half-true ruling. The summary comments that it leaves out details, but enough of what they think those details should be are there in the statement.

        And thats why people are questioning their bias. Not the rulings, but the ratings.

        • Not the rulings, but the ratings.

          Exactly…if they left the ratings off, they’d actually be pretty good.

          I actually prefer the ABC’s rating system as it has the whole “shades of grey” thing built in, rather than trying to be “black and white” about it.

          http://www.abc.net.au/news/factcheck/

  17. don’t worry the coalition is just about to take a huge dump on the NBN so these little smoke screens, distractions are irrelevant. What is relevant though is this:

    What could’ve been, you will never know.

  18. The problem with Politifact OZ is that they are shit.

    Their modus operandi is to focus on semantics and pedantry instead of the substance of any issues it “investigates”.
    As highligthed above, we then get rulings on “gold-plated coffee machines” where they are actually looking for coffee machines made of gold.

    Basically they are yet another media beard for the conservatives, attempting to cloak themseleves in the aura of respectability of the parent company they licensed the name from.
    If anyone from Politifact in the US reads their site they should sue them for misrepresntation.

    • Except this article is an example where they didn’t focus on semantics and pedantry.

      If all they did was focus on semantics and pedantry I’d dislike them, but I wouldn’t call them wrong, I’d call them pedants. If they could be consistent with their ratings I couldn’t complain about them.

      But they *are* inconsistent. That is what is wrong with politifact, that is why they are – as you say – shit. Not the pedantry, not the semantic arguments. But the inconsistency.

      • Like you said the problem is the inconsistency…in particular inconsistency to achieve their aim of making one side look bad.

        They use all of their tools to get it done, like semantics and pedantry on the ALPNBN ads, then cut Malcolm several miles of slack with his costings bullshit in order to give him a faint cloak of respectability with “half true” when the more appropriate rating is “steaming pile of crap”.

        For the unaware out there PolitiFail Oz is a trap because they will automatically assume because it purports to be “fact checking” that it is being even handed and consistent.

  19. Michael

    “What a tangent. All I said was that irrespective of your data, the same methodologies is used in analysing it.”

    it is obviously difficult for you to understand that there is no point using any methodology to analyse data if you know the data is flawed. Evidently it is beyond statistics 101.

    “Can you find a definition to support your view that they are only large”

    I don’t need to because this is not what I said. Read carefully. There is a population of statements made by politicians. In this instance, politicians in Australia. If a few of their statements are selected for an assessment of their veracity, in an arbitrary fashion, it is not longer a random sample of that population. As result, that sample may not be representative of the larger population and, therefore, the findings are quite possibly meaningless.

    “A population can easily be part of a larger one”

    And to emphasize my previous point, it, then, needs to be selected randomly which this sample was not.

    “I said that they had more results on the page in Politifact.”

    No, you didn’t. This is what you said

    “Ahh the irony when you are calling him biased for reporting on a quantitative statistic.

    It does show a logical fallacy and another measure of bias.”

    A very pompous, yet meaningless way of showing off you limited understanding of research methodology.

    Had you said what you now claim you said, I would have replied that the results on the Politfact’s page were chosen arbitrarily and therefore may not be representative of all the statements made by Australian politicians. Furthermore, I would have, as I have previously, pointed out that given the shortcomings of the methodology and of the ratings, it is irrelevant which party is deemed to have more false statements, given the way truth and falsehood can be defined in degrees in the ratings.

    So to conclude Fibroid did not report on a quantitative statistic. What he did was report a specific finding, claiming that his choice was warranted, given the greater number of reported false statements by Politifact.

    • Who ever claimed that the politifact findings constituted a sample of the total population? Let alone a represenative sample.

      Nice non-sequitor.

      • This was my best effort trying to answer your disjointed argument.

        Have a lovely night.

        • I suppose it was futile from the start.

          You do not respond to the topic, and repeatedly go off on tangents.

          I will leave you with the previous question as one of many you declined to answer.

          “Who ever claimed that the politifact findings constituted a sample of the total population? Let alone a represenative sample.”

          • @ Michael… the other day you asked “which threads Renai had to close down the most”…?

            http://delimiter.com.au/2013/08/16/treasury-should-cost-coalition-nbn-policy-labor/#comment-620713

            To answer your question here, since that thread has come and gone…

            IMO, having seen many posters (myself included) involved in futilely trying to explain to a few unconditional NBN detractors, what to everyone else seems to be the bleedin’ obvious, … I’d suggest the threads closed down the most are the NBN threads where these few 24/7 naysayers lurk. Where they Intentionally endeavour to derail NBN positive and fully elaborate upon NBN negative articles, seemingly just to antagonise poster, with what in reality is nothing more than childishly argumentative, nit-picking bullshit…

            Much like is happening here…

          • I think you and I are on the same page. Arguments all over the place, avoiding major points and always ending insisting on an answer to an irrelevant question.

          • In my opinion the pro-NBN and anti-NBN diehards are as bad as each other, and I would prefer neither commented on Delimiter. I would prefer this forum remains a haven for open-minded people capable of changing their mind on issues when sufficient evidence is presented to make them do so, and I am currently moderating comments with this aim in mind.

          • I believe it depends on where becoming a diehard lies.

            If you are pro or anti NBN because you have done your homework and truly believe one option is clearly better than the other, then so be it.

            But if one’s loyalty either way is purely political, well…

            Seems to me the NBN is very emotive and if you took the diehards (and yes, I am a pro-NBN diehard, NOT for political reasons)… there wouldn’t be many people left to post.

          • The issue is what you do when evidence is presented contrary to your opinion. If you don’t at least open your mind to the evidence and consider it, then you’re not the sort of person I want commenting on Delimiter.

          • The problem with much of the debate, about the two competing plans, is that there are many unknown in the Coalition plan. This leaves it open to assumptions and interpretation. In many cases, evidence, per say, is hard to come by, unless you have a crystal ball or have a lot of faith in politicians.

            I personally have no issue with the Coalition plan being achievable. My concerns are whether it can be done as easily, effectively, accurately, economically and quickly as they profess.

            I do have issues with people from both sides who see it as Labor vs Coalition. Why? Because I think that such an important infrastructure should have bipartisan support. If it had, you could have the best possible network with input from both sides, instead of an ideology driven one.

          • I think you should just take the higher ground and walk away sometimes (usually once you realise that “Poster X” would rather die in a ditch than concede his point).

  20. If, as Politifact say, the $94 billion claim is unverifiable then why did they bother to “fact check” it?

    Until it is actually built, the claim can never be proven true or false so, in that respect, the exercise is pointless.

    The issue is not whether the $94 billion projection is true or false, but rather whether it is likely or unlikely.

    On that point, Politifact says that it’s possible.

    Politifact should have elaborated on this further by making an estimate on what they think the NBN would cost. That would have put things into better perspective for readers.

    • “Politifact should have elaborated on this further by making an estimate on what they think the NBN would cost.”

      Riiight, because that’s a possible thing for Politifact to do, given their level of resources.

        • @ seven_tech,

          If Politifact came to the opinion that the NBN would be a total cost of $44 billion rather than $95 billion then that conclusion would lend support to Labor’s position. A rise of $1 billion in cost isn’t that much of a disaster.

          • @Kingforce

            Yes, but that’s not really the point. Politifact have nowhere near the resources to estimate and test those assumptions on either side. Like I said, if the PBO can’t do it with millions more in funding, what chance do they have?

            Politifact are an organization based around fast turnaround and specific point analysis of political claims. Testing the assumptions & modelling for the NBN fall well outside that.

          • Politifact don’t need to commission a study that costs a lot of money. They just need to use the opinion of respected industry sources to give the public an idea of what a possible increase in cost might look like.

          • @Kingforce

            True. But there’s not really much point considering that’s not their stance. They usually have 1 or 2 contributors that do that for the article. Besides, several high profile industry analysts have already made their assertions about the $94 billion….

      • The public needs to be some context. Can’t they interview a range of independent industry analysts? Would have thought that would be the obvious thing to do.

Comments are closed.