Monday, 13 May 2013

NBN: The NBN Co business model - who really subsidises others.

There are two counterfactual memes, or wrong Urban Myths, running around:
  • Why should I pay for something, like the NBN, that others will get the most benefit from?
  • My connection is just fine, why should I care about the NBN, I don't need it?
Early users, especially those signed up for high-speed (100Mbps & 1Gbps), pay higher rates and they subsidise the rest of us, not the other way around.

At 30% growth rate in download volume, the conservative ABS figure used by NBN Co, every 32 months data download doubles. Instead of paying twice as much, users will only pay 14.6% more. That's a really good deal, I wish I could have that for petrol and electricity as well.

NBN Co intends to halve volume use charges every 4 years: for each 30% growth in traffic, they'll drop the price-per-GB by 19%.

It's not been spelled out clearly anywhere I've seen, and is even given prominence, as a negative, in the Coalition NBN policy on the main page of their policy. When it is in fact, exceptional value for money: a three thousand-fold decrease in $$/GB.

By contrast, under Labor’s NBN wholesale charges per user will triple by 2021.
NBN Co, 2012-2015 Corporate Plan, p.69 and 72. Page 69 of NBN’s Corporate Plan displays a graph showing a steep rise in the monthly ‘ARPU’ (average revenue per user) NBN Co expects to collect. NBN Co’s financial projections on page 72 show wholesale charges per user increasing from $22/month in 2011-12 to $62/month in 2020-21.
The growth in NBN Co ARPU charges are from data download volume, not access charges to the service. This is discretionary expenditure by customers and based on their willingness to pay. If subscribers don't want to pay more, they don't have to. What we've consistently seen in nearly 20 years of commercial Internet in Australia is that:

  • Australians love new technology, they are always in the forefront of its adoption, and
  • We back-up our interest with cold, hard cash. Telecommunications has always been immensely profitable in Australia with very high elasticity: if you drop the price 10%, you'll increase sales and profits by a much, much higher amount.
For average user charges, ARPU, to triple when doubling download volume only increases charges 14.6% means average downloads increase 3,100 times. Over 11 doublings, three thousand times more data for 3 times as much money. That's not just a good deal, it's a fantastic deal.

Secondly, you need to understand median and averages.
The median wage of taxpayers in Australia is just $45,000/year: this is the wage that half of us earn more than, and half less than, the midpoint.
The average wage of Australian taxpayers is $66,700 (source: ATO via Peter  Martin).

That extra $21,000 is effect of huge wages of the top 1%, dragging up the average. [correction:Tks to Paul K]

This is the same deal with NBN Co. They've forecast most people will stay on the lowest-speed plans, and accordingly on lower monthly usage volumes.

Like wages, there will be a very few people running the very highest speeds and downloading massive volumes of data. These are the people who subsidise the rest of us. They pay more, we pay less - that's how averages work.

NBN Co doesn't base their forecasts on median, or midpoint, data volumes, but average data volumes.
The top 1% of customers may easily account for 50% or more of total data downloaded: this is a very good thing for the rest of us. Those Big Users bring the price down for the rest of us, and not by a little. We're talking thousands of times in the next 10 years.

For those not familiar with the effects of Moore's Law, these claims might seem outlandish.
Think of the first IBM PC of 1981. Current PC's have a CPU at least 10,000 times faster, the disk a million times larger and the price less than one tenth. Sales volumes are up by 1,000 times or more as well. This is the curve that NBN Co are wanting to drive us down.

That's the Real Deal: a tripling of the ARPU by 2021 is a Great Thing.
It means for most of us on minimum plans, we'll pay a whole lot less or get hundreds of times more data download for the same money.

Instead of abusing the product leaders for diving into the new technology, we should be applauding them. If they want 10Gbps services in 3 years time, give it to them! Every $100 more they spend is that much less that you and I need to spend.

Answering people who say a version of "I'm alright Jack" there's some simple questions to ask them:
How many times have you upgraded your Internet connection since 1996? How much faster is your link now compared to then? How much less per GB do you pay now? How much of your life now depends on or is facilitated by the Net? Did you foresee any of that?
Which leads to the crunch question:
If you didn't plan or foresee any of your increased Internet usage over the last 15+ years, why do you think your use of and reliance on the Net will stagnate now?
Because change is gradual and few people carefully track costs and usage, most people haven't noticed the degree of change of their normal home experience.
For them it's like electricity, airplanes and gas: It Just Works, what's the Big Deal?

The Big Deal is that we've already increased our Internet usage 1,000-fold, with quite modest increases in real-prices. It's there in multiple long-term data series of our average monthly data volumes, they're still increasing at heart-stopping rates, at least 30%/year.

The kicker is that the ordinary folk who've already got onto the NBN are using the Internet at 50% higher rates and 8 times as many as predicted have signed up for 100Mbps. The data say that the NBN Co forecasts are low, that actual demand, and so revenue, will run far ahead of their projections.

For those who are happy with what they have, I don't question their position if they are actually aware of their change in usage over the last 15 years.

To the rest of them, I ask:
If a decade ago you didn't foresee what you do and take for granted today, why do now claim to know what you'll want and use in another decade?
"I can't see it" and "Why would I want to do anything more?" are failures of imagination, not informed statements.


  1. That extra $11,000 is effect of huge wages of the top 1%, dragging up the average.

    Should that not be ~$21,000? (66,000-45,000)

    Feel free to delete this if im right, and definatly if im wrong :)

    Thanks for your artticles, great analysis as always.

  2. Paul,


    Yes, You're right. guess I my Mental Arithmetic Unit has gone flaky like my memory :-)

    Joys of being Slightly Older.

    Glad you enjoy my pieces.


Note: only a member of this blog may post a comment.