NBN Revenues Provide Interesting Calculations

There were some very interesting revenue figures released as part of NBN Co’s Annual Report 2012 last Friday.

“NBN Co has reported revenues of AU$2 million for the first year of commercial operations for the 13,536 premises connected to the National Broadband Network (NBN), ending June 30, 2012.”

The NBN when completed is designed to cover approximately 12 million premises. If NBN Co can deliver $2 million in revenues over 13,356 premises, lets do some simple maths with the currently factual numbers.

Presuming a 75% uptake of NBN services across the twelve million premise footprint – (ie: nine million premises take up a service) – if 13,356 premises deliver $149.74 each – (to total two million dollars in 2012) – then nine million premises – (at the same rate) – would deliver a revenue flow of $1.34 billion.

Furthermore, it is likely that the annual per-premise revenue will be higher than $149.74, because many – (in actual fact, most) – of those 13,356 premises would not have been connected to the NBN for the entire 12 months covered by the annual report.

Given that the base wholesale NBN pricing is $24.00 per month, theoretically, the minimum annual per-premise revenue should be 12 times that figure, or $288.00 – which is more than double the figure in the annual report.

Nine million premises, at $288.00 per premise gives a revenue flow of $2.52 billion.

Not bad for a “white elephant” that will never pay for itself! It should also be pointed out that NBN Co didn’t start charging ISPs until October 2011.

I also suspect a lot of people will say something like “75% uptake? They’re not even close to that!”.

Sure. That is true – right now.

But as Telstra gradually decommission the copper network within the NBN fibre footprint, people on the existing copper network – (even if it is just for a telephone service only) – will necessarily need to be transferred to the NBN by their service provider to continue that service.

Given also that it is possible to have more than one service per premise, it is technically possible for take up to go beyond 100 percent.

So 75% might actually be a very cautious estimate.

It is making money, and based on these current revenues, a completed NBN should be making a lot of money for a long, long time.

  • You Zealot. how dare you publish such controversy!

    Seriously thou. do you think that the common man knows this. Or are that listning to Tony Abbot and the main stream media that publishes articles like “NBN Co lags behind schedule while bosses get ahead on bonuses”. Its hard to know what the general public believe.


    • I look to publish facts – not the “facts” that some politicians believe in… 😉

      Spending money to advertise some facts is perfectly reasonable…

  • Jim

    Michael, i think you’re seriously confused about this. It is wrong to extrapolate that $2 million revenue can be divided by total customers to give a yearly revenue spend. Many customers will have only connected to the NBN for part of the year. $149.74 per customer per year is extraordinarily bad — that assumes an ARPU of just $12.4. On p.69 of the current corporate plan, the NBN needs at least $20 ARPU in the current years and then push it up to $60 by 2020. The NBN are providing very little information on revenues but as you can see by the graph on p.69 they’re below the 2010 projections on ARPU. No offence but this was one of the most intellectually weak analyses of the NBN economics I have ever seen. If you want to follow someone who knows their stuff, read Ian Martin who publishes often in the TJA.

    • Jim – I actually said that in the post. That most, if not all, would only have been connected for a part year. I also pointed out that NBN Co didn’t even charge for the full year, starting in fact in October 2011.

      This is also why I provided the theoretical figure of $288.00 per customer service – (12 x $24.00) – as a comparative measure.

      In using the $149.97 figure, I’m pointing out that even it’s infant stages, when pricing wasn’t even place for the whole year, that it still makes a revenue return.

      If you’re trying to suggest that $2.52b – (or even $1.34 billion based on the lesser per premise result) – is not a decent amount of revenue, I don’t know what you think is.

      ARPU is not the guiding principle as to NBN Co’s revenues. Their goal is to return the investment, with a 7% return, over the 20 year life of the project.

      They’re not their to “make money” – ARPU is therefore irrelevant. $24.00 returns their costs, plus 7%.

      Nothing more, nothing less.

  • Jim

    But by 2021 (when the build is finished) they’ll need $6.1b in revenue pa to be on track to making a return (corp plan p.72). So no, $2.52 b is not impressive — it would mean the gov is on its way to burning 58% of its investment.

    • You seem to be of the mistaken opinion that the project is a ten-year project.

      The build project is ten years, but the financial terms of the project run for 30 years. If it were 10 years, your $6b+ figure would be close to the mark.

      But it would also mean the approximate tripling of the wholesale cost – ($24.00 vs $72.00) – which would defeat the goal of maintaining rough parity with current pricing levels.

      Some ISPs still fork out more than $24.00 to Telstra for wholesale access now, so $72.00 would be a laughable figure.

      No, the financials terms are 30 years, so your point is completely invalid.

      • Jim

        Wow. Just wow. I don’t know where to start.
        Alright, first point — it’s a 10-year project. Remember the policy as conceived by Labor is to sell the NBN once the network is complete so the revenues in 2021 are incredibly important.
        Second, you are denying that the $6b+ figure is correct. That is what the NBN themselves have said in their corporate plan — see p.72 http://www.nbnco.com.au/assets/documents/nbn-co-corporate-plan-6-aug-2012.pdf
        You simply can’t dispute that the $6b+ figure is needed as conceived by the NBN to recover costs. By the end of the 30 year period, by the way, the annual revenue needed to recover costs is more than $14b pa. Remember the full extent of the recoverable costs includes both capex and opex — that is, they have to recover all costs of Telstra, Optus deals etc. So don’t try to backsolve projected revenue streams purely on the $37.4b capex. Also, we still don’t know how the NBN quantified its Initial Cost Recovery Account in its SAU — but it has basically been rejected by the ACCC anyway and they have had to go back to the drawing board.
        As to tripling of the wholesale costs — that is exactly what they are proposing to do. See the proposed uplift in ARPU on p.69 of the corp plan.
        In terms of prices, the NBN may say its planning to hold prices on specific products at half of CPI but it is also proposing to a) impose CVC costs to recover revenue through increased usage (no wonder they love quoting the ABS stats on usage) b) upsell customers to other products like multicast ($20 a month) and c) upsell customers up speed tiers.
        And finally, you say they have exceeded a $24 ARPU in 92 per cent of premises — my whole point is that you can’t simply equate AVC charges with APRU. You’ve clearly never done any financial modelling for a telco before.
        So to recap, the whole premise of your post is just utterly, embarrassingly wrong. The assertion that $2m in revenue shows that the NBN is now a great cash-cow for taxpayers is just wrong. It shows us … nothing of great significance. We simply need much, much more data to make an informed opinion on the economic viability of the NBN project

        • You could start with accepting that you’ve gotten it wrong.

          The BUILD project is 10 years. The FINANCIAL aspect of the project is 30 years.

          The outlay is set to be recovered over THIRTY years.

          Since you’re so fond of quoting the business plan, check out page 9, and I quote:

          “The higher level of connections and revenues beyond FY2021 results in higher projected cash flows, which increase the forecast return (IRR) from 7% to 7.1% over the 30 year period used for the NBN Co financial model (from 1 July 2010 to 30 June 2040).”

          Sounds like thirty years to me.

          Secondly, I *NEVER* related anything to ARPU. You know, when I said “They’re not there to “make money” – ARPU is therefore irrelevant.”

          It was YOU who related it to ARPU when YOU said: “On p.69 of the current corporate plan, the NBN needs at least $20 ARPU in the current years.”

          Am I denying you $6b figure? Yes. It is easy to do so. If all 12 million premises took the top plan at $34.00 pcm, that’s only a smidge over $4b.

          Yes, that doesn’t include CVC pricing, but that’s near impossible to model, because every ISP will do it differently.

          If you can’t accept that the project has a 30 year financial life, it’s clear you haven’t read the business plan you keep quoting from – rather, you’ve cherry picked the pieces that suit your argument.

          You argument is therefore, invalid – because you are basing it on inaccuracies – deliberate or otherwise.

          What is embarrassing is your refusal to believe the 30 year project length, and your attempts to wriggle out of it, while still refusing to accept it.