Why NBN is so expensive

Techradar has reported that new cheaper NBN plans could be on their way before Christmas following consultation between NBN Co and the Australian Competition and Consumer Commission (ACCC):

While consumers may think the problem of pricing lies with the ISP they’ve signed up with, the issue starts with NBN Co, who charge the providers a wholesale price. The proposal looks to lower that wholesale price which, in turn, will allow retailers to pass on cheaper plans to customers…

NBN Co’s plan follows extensive consultation with the ACCC, and looks to lower the wholesale price of the NBN, which will allow retailers to pass on cheaper plans to customers…

It’s not yet clear how a new pricing structure would benefit the end user, but the proposal includes changing the rebate from a one-off payment to a daily rate, which consumers will be eligible for when they face late connections and fault repairs…

If the proposed changes are approved by the ACCC, they’ll come into effect by the end of the year, giving more Australians access to affordable high-speed internet.

These changes, which haven’t even been properly defined, amount to tinkering at the margins and do not resolve the fundamental structural reasons behind the NBN’s exorbitant wholesale pricing.

The NBN’s wholesale prices have been set at a high level because the former Rudd government classified the project as an “investment”, which necessarily requires it to deliver a commercial return. This meant the NBN must cover its costs in addition to earn a margin on top, which necessarily forces it to charge ISPs high wholesale prices, which are then passed onto Australian consumers.

The ultimate solution for the NBN’s high cost is for the federal government to write down the value of the project to reflect its true value.

The Parliamentary Budget Office recently reported the “fair value” (or saleable value) of the NBN at just $8.7 billion, which is less than one-third of the government’s equity investment. By extension, the NBN requires a writedown in the order of $20 billion.

Writing down the NBN would lower the project’s required rate of return and enable NBN Co to lower wholesale prices for ISPs and by extension Australian consumers.

Ultimately, the federal government needs to treat the NBN as an essential utility service, rather than a commercial project seeking a commercial rate of return.

Until it does so, Australians will continue to be overcharged for substandard broadband.

Unconventional Economist
Latest posts by Unconventional Economist (see all)


  1. Jumping jack flash

    “Ultimately, the federal government needs to treat the NBN as an essential utility service, rather than a commercial project seeking a commercial rate of return.”

    This 100%.

    The alternative is that since each of the employees at NBNCo. will need to be paid an income that makes them eligible for the enormous amounts of debt that are required to obtain and maintain their expected standard of living, this wage cost is passed onto the consumer, plus a bit extra to provide the expected amount of profit and growth. This makes the prices higher.

    Or, they could substitute in 3rd world slaves wherever possible and then steal wages from them to try and keep the prices down, but that’s not completely necessary in this particular scenario. Quite simply, the sky’s the limit for an essential utility service.

    The same model is used by all privatised essential utility services, as we currently experience.

    • But the “commercial project” was how Rudd hid the impact from the Cth budget. If it isn’t “off budget” – all those costs go through the budget each year – hurting Joshie’s hopes to ever get back in the black.

      Now the main problem is that the NBN is so expensive because they didn’t just do Melb Sydney Brisbane and started with marginal electorates like Tasmania and Windsor’s tamworth seat first. It has been non-commercial from the get go.

  2. Display NameMEMBER

    There is no such thing as a public good? Everything has to have a commercial return ?
    We know the price of everything and the value of nothing

    • The commercial return was ALP’s way to hide the budget cost – so they didn’t need to face scrutiny for it. Libs have merely kept the trick.

      The Dept of Defense doesn’t have a commercial return – but that is why the cost of the subs is front page news all the time. This is why the Govt hid the NBN off the books.

      • That’s just not true though. While it wasn’t on-budget, *exactly* how much the Government was putting in, and exactly how much was being spent was disclosed every year in the NBN Corporate Plan, as well as other financial updates that would happen from time to time. There was absolutely every opportunity for scrutiny. Funnily enough, Turnbull had these accounts forensically audited when he got in, because he couldn’t believe the NBN was on-budget at that point (it was!). They loudly and gleefully announced the audit, but when it was done it was extremely quietly released because it found “no material issues” and they wanted to keep up the narrative that it was over budget and needed their changes to “save” it.

        None of the secrecy really came about until Turnbull removed the whole board and management and put his associates in place. After that, annual reports became massively redacted (they went from ~300 pages to 40-50ish), and far more details became “commercial in confidence”. That was extremely hypocritical because one of Turnbull’s chief complaints (which was a lie at the time anyway) was that he said NBN Co wasn’t transparent enough. And he just had so much shrouded in secrecy once he was in!

        In reality, the reason it was off-budget was because they worried that was the only way they could get it through past the neoliberals and without a massive scare campaign about “socialism” (as if the Government building electricity distribution or the original Telecom copper network was any different?). Even as it was, Turnbull (stupidly) called it a “Soviet” style program.

  3. It’s pretty simple to work out why.

    Average connection cost / years to pay back = minimum required ARPU over the period.

    Let’s say it’s 50bn / 10m connections / 20yrs = 250/yr

    = $20.8/m barest minimum ignoring operating costs and interest.

    • It is not something that should be built to receive a direct return on investment. The purpose of the NBN should have been to provide the best internet possible to as many people and businesses as possible at the lowest possible price. This would require the connection and usage fees to be lower than the amount required to cover the cost of the build.

      Why would a government do this? So that the infrastructure can be used to reduce the costs of business, improve productivity and allow for the uptake, and ideally early adaptation to a technology that is fundamental to life and business in the 21st century. The whole of the community subsidizes it and the whole of the community benefits from it.

      An international example of the is the European Global Navigation Satellite System, Galileo. It’s improvement in positioning, timing and safety of life has been calculated to value add approx 100 billion euro a year. In fact the European Space Agency is a model for this as they provide a lot of data for free. This has driven research in the remote sensing field which improves our understanding of the world and leads to gains in agriculture (crop and livestock management), environmental risk analysis and many other things.

      Small shopkeeper thinking at the governmental level may appear like good management when things are going well but it inevitably ends up stunting the growth of the nation and eventually turning it into a pauper.

      • Maybe so.
        But neither you or I make that decision to write it off as general government expenditure.

        Until that changes, that’s the maths.

      • I would agree with the comment below it it was true;

        “…The whole of the community subsidizes it and the whole of the community benefits from it.”

        – Businesses for the most have been able to pay for and write off the cost of a fibre optic connection if needed to conduct business. Most Telco’s offer these connections and most businesses who need it have the in office technology to maximise its speed (High performance switches and routers, high speed enterprise wifi (expensive) or ethernet to the desk).
        – Not all homes need the capability fibre provides. Even if you connected FTTP to every home (massive cost) majority of people would waste it with inefficient last hop wireless setups.
        – It’s been a $50B cost with a stripped out program of work. Imagine if nearly all urban homes actually got FTTP, the cost would be astronomical. A mini excavator and a crew to dig up peoples front yards runs into the thousands each day.
        – Only some homes are using the connection to generate a profit and grow the economy. Not everyone is a knowledge worker and many use the internet purely for entertainment and recreation purposes. This is a 100% legitimate purpose but does Netflix, Stan and Foxtel viewing warrant a $50B cost?
        – Urbanites gain the majority benefit while rural and regional areas continue to subsidise without a direct benefit (Galileo is not a great example because all users can access it equally)
        – It was never rolled out equally due to its highly politicised nature. Some locations got much better connections earlier than others later in the program (FTTP vs FTTC vs FTTN vs Fixed Wireless).

        The only way the NBN would have worked is a guaranteed minimum speed per connection (tech agnostic). Above the nationally regulated minimum it should become a “user pays” system. Instead it became a technology argument slanted by political bias on both sided. We should have discussed outcomes 1st, technology 2nd and then regulated accordingly to give flexibility over delivery.

        • The only way the NBN would have worked is a guaranteed minimum speed per connection (tech agnostic). Above the nationally regulated minimum it should become a “user pays” system.

          You have just described the reasoning that lead to the already-semi-obseleted clusterfvck of Turnbull’s MTM NBN.

          In reality we’re almost certainly going to end up with FTTP (or it’s nearly identical twin, FTTC) anyway, in another example of “doing the right thing after exhausting other options” (after which it will probably be sold off, in an example of “snatching defeat from the jaws of victory”).

          • Great call, that write down will occur just in time for an unexpected commercial buyer to step in, ready to selflessly assist tax payers with our asset recycling needs while simultaneously delivering more efficiency, better services and a healthy corporate profit margin…. ah the miracles of modern economics!

        • Imagine if nearly all urban homes actually got FTTP, the cost would be astronomical. A mini excavator and a crew to dig up peoples front yards runs into the thousands each day.

          I don’t want to take away from the value of your other comments, but lets consider that bit about the excavator. You have assessed the cost at thousands each day.
          I assess the cost as the manpower to crew the mini excavator and the manpower to build and maintain the excavator. Maintain includes producing the fuel to run it.

          Let us say that 2 men are needed to operate the excavator and that for every 2 men operating excavators there is 1 man working “behind the scenes” to build and maintain and fuel the excavator fleet. This is the total resources consumed by the activity.
          So to dig up the front yard of your house would take 3 men 1 day including all costs. That seems cheap to me. I would be prepared to pay 3 days of my wage for that FTTP installation. Cheap.

          What is astronomical about that? What have I missed?

          • What’s astronomical is the shear number of small fiddly connections. Machines run at about $100-$150 an hour, they have to call the power companies, gas companies (dial before you dig). You then need shovels close to cables/pipes (assuming they have been marked properly). You then need the technician to do the actual connections (probably won’t be the person driving the digger) and most likely a stop go person for busy roads. All has to be OHS approved so things take longer than expected.

            You haven’t even left the depot and your over a grand. On new builds where the fibre goes in at the time of excavation work the economics are completely different.

            Many other countries did not have the technical debt, copper along with along with very aged housing of Australia

          • > So to dig up the front yard of your house would take 3 men 1 day including all costs. That seems cheap to me. I would be prepared to pay 3 days of my wage for that FTTP installation. Cheap.

            If you’re in an FTTC area, you can probably do that if you really want fibre.

            Not many know, but FTTC has a direct upgrade path to fibre – which is expensive for the first of the four connected to the box, but if you can convince your 3 neighbours connected to that box to upgrade is cheaper. It only requires running 4x fibre lead-ins, external box, internal fibre, adding modem, and replacing the ‘node’ with a fibre splitter box.

            If the conduit is in good condition, 2 guys could probably do all 4 in a day.

          • Not many know, but FTTC has a direct upgrade path to fibre […]

            Not many know ? Isn’t that pretty much the whole point of FTTC ? 🙂

  4. And since the NBN is an essential public utility (and a de-facto monopoly) it should NEVER, NEVER be privatised but remain publicly owned; that is the only guarantee the interests of AU’s residents are served and not those of rent-seeking shareholders.
    Anyway, if the true value of the NBN is AUD 8 billion = the monetary equivalent of just 8 weeks of Covid-19 emergency support; selling the NBN for just that money would be the stupidest thing ever and set back AU in time another 20 years to the time where Telstra’s monopoly strangled progress and kept prices ridiculously high.

    • In that case, it will be sold for 5bn and be heralded as a success. Julie Bishop will join the board of Telstra and life will go on as normal in Oz.

      • Yes, you mean the Julie Bishop who can’t deal with figures 🙂 (she has a reputation of being not really good in number crunching); if the NBN gets sold we might as well sell the ATO, the Army, Medicare, well even auction off the title of Queen.

        The tragic bit is that several states already have committed similarly unwise acts by selling toll rights, Land Title registration offices, power grid lines. These acts will become the future economics textbook illustrations what NOT to do 🙂

    • “What is the value of the NBN?”

      It allows citizens to access the Internet faster than ever before.

      “No no no. I mean what is the monetary value of the NBN? What would it sell for?”

      Well there is the NBN hardware (copper and fibre lines and modems, powersupplies, buildings) and there is the licensing arrangement with govt.
      The NBN hardware might sell for $1 Billion if ripped-out and shipped to say Indonesia.
      The NBN hardware might sell for $8 Billion in a competitive market situation for Internet supply. i.e. If the NBN was destroyed, in a competitive market situation a marginal supplier might pay $8 Billion to recreate the functionality of the NBN.

      However if the NBN is sold under a monopoly licensing arrangement from govt, then it is worth the maximum that Australians are prepared to pay for Internet access. It might be $100 Billion or $800 Billion.

      Sadly here in Australia our rulers count themselves as good economic managers by selling assets attached to monopoly licensing arrangements.

      • You are completely right Claw, however, ANY politician who intentionally sells the NBN as a licensed monopoly to a private rent-seeking shareholder (e.g. Telstra) _without_ strict Gov set price controls should be charged criminally IMHO.

  5. SnappedUpSavvyMEMBER

    got till November to move over to the freakin nbn, the 80mbs plans are expensive! is 60mbs any good for, you know, watching ‘movies’

    • darklydrawlMEMBER

      Got the same problem. Been on Telstra cable for years getting a stable and reliable 120 mbps download. The best Telstra can offer (due to the FTTC) is 50 mbps for the same cost. Not Happy Jan! The only potential upside is the uploads should be faster on the NBN – assuming the 40 year old copper wire is up to it.

      I have looked at getting an external provider for internet. I can get a direct / dedicated microwave link @ 75 mpbs (up and down which is decent) – but it is not cheap. Hmmmmm…..

    • I get by on 10 MBS down and 1 MBS up. It’s a single person household and I stream a lot through a VPN.

      If I’m working from home I occasionally download and upload large datasets, but there has never been such a tight deadline that I’ve ended up in trouble.

      I don’t game so I can’t comment on that.

      The one thing I’d note is that fttp has been more consistent and more reliable than my previous connections. It will be one of the must haves of the new place whenever it is I decide to switch rentals.

  6. Instead of building the NBN, we should have just bought 51% of Telstra, deregulated them, and let them build it. We would have had a profitable organisation from the beginning.

    • darklydrawlMEMBER

      Or you know… Maybe never have sold Telecom / Telstra in the first place. Sh!ts me that the taxpayers already owned all of it, and now we have to pay again. Idiots! Greedy swine the lot of them.

      • Haywood JablomyMEMBER

        Taxpayers build a network which is then sold to taxpayers, who then pay again to buy the sh!tty old network that they paid for and sold to themselves, then pay again to add good bits to the sh!tty old bits, then sell that at a knockdown price to the network owner that they sold to themselves.

        Seems perfectly sensible.

  7. I’m enjoying the NBN with FTTP at 250 Mbps down / 100 Mbps up – the greatest and most stable internet connection I’ve ever had. It shows the rest of the nation would greatly benefit from having their crappy FTTN copper last-mile upgraded to proper FTTP at one point in the hopefully not-too-distant-future 🙂

  8. HFC is OK speed-wise but suffers from outages much more than FTTP – so that will remain a balanced decision, but if you have choice going for FTTP is preferred.

  9. UpperWestsideMEMBER

    In NYC I have HFC , Cable modem -> google mesh wifi -> google mesh wifi -> apple laptop
    Ookla gives me 138megs download (average, peak is 148megs) 15 up. Speedtest from the first google mesh wifi to a google node is 220 megs download , 15 up. As I only pay for 200megs that seems fair. Top speed they offer on HFC is 980megs download.
    So as a tech, HFC works fine.