NBN Based Internet Industry in Strife – an update

Some months ago I wrote a post expecting some providers would get into trouble as their margins are being squeezed between NBN price rises and the unwillingness of the telco industry to raise their prices (and the market to accept them).

Since then two small providers, Telecube and Mungi, have gone bust and another, Amaysim, has exited the market claiming there is no margin to be made. Then there is Vodafone Australia who has decided to join forces with TPG rather than continue on it’s own in the NBN space.

To understand what is going on here, it is important to understand that there are different ways to provide services on the NBN. Most small providers want to have national coverage – i.e. be able to sell internet services all across the nation. Due to the NBN 121 POI model, it is prohibitively expensive to connect directly to every POI just to get started. So small providers take the only alternative which is to go through an aggregator. An aggregator is a large telco who has paid for the physical fibre access to all the POIs and agrees to on-sell capacity in the form of aggregated bandwidth across all the NBN POIs to the small provider. The aggregator simply charges an access fee per connection and a price (per Mb) on the sum total amount of bandwidth allocated to the provider’s clients.

However there are four big downsides to this method:

  1. The small provider has little or no control over the congestion at each POI on the upstream aggregator’s network – the provider’s traffic is lumped in with all the aggregator’s other customers including possibly their retail arms. It is very hard for a small provider going through an aggregator to promise a “high quality” network when they don’t control the part that creates most of the congestion (the connection to the POI – the CVC).
  2. There are only a small number of aggregators in the market so competition is not strong and prices for the bandwidth are quite high (of the order of $15 to $20 per Mb – including NBN’s CVC charge).
  3. The small provider can really only offer products that their aggregator offers. They cannot, for example, offer higher priority traffic or more bandwidth per connection unless their aggregator chooses to. This limits the ability for the small provider to innovate and it becomes yet another “me too” provider competing on price. A good example of this is some aggregators unwillingness to provide the NBN’s new (cheaper) bundled plans, something Mungi blames for their collapse.
  4. The small provider has no direct access to NBN and NBN’s management systems. Everything has to be done through their upstream aggregator. If there is a fault for example, the client calls the RSP, the RSP calls the aggregator, and then the aggregator deals with NBN. Some of the automation is very poor in the aggregators, performing what has jokingly been called “swivel chair management”, where the aggregator receives an order through one system and manually retypes it into another system. This is not efficient, leads to errors and unhappy clients.

At Launtel
We got so fed up dealing with the above we decided to eschew the aggregators and connect directly to NBN. We have had to give up (for the moment anyway) national coverage because of this. Aussie Broadband I understand found exactly the same issues with their aggregator and decided to embark on an ambitious (and by all accounts expensive) project to connect to all 121 POIs.

So in a nutshell, the small provider generally has higher costs and less ability to innovate in the market (i.e. create more value) due to their use of aggregators. It is worth doing some back-of-the-envelope calculations to see what these small providers are up against.

Let’s say they want to compete with TPG’s 50/20 unlimited download service, currently priced at $69.99 including GST. To provide a low congestion, unlimited download service you must allocate at least 2Mb per connection (on average). It is debatable if this will be enough long term, but we will use it for now. Let us assume the following monthly costs (ex GST), note we are not using the bundled pricing, because few aggregators seem to be using it:-

Item Unit cost Line cost Total
NBN 50/20 connection charge (must be at least this) $34.00 $34.00
Aggregator per Mb charge $15.00 $30.00
ISP Network/Peering/IP-Transit $5.00 $10.00
Billing, support, admin $10.00 $10.00
Sub Total $84.00
Add 5% profit (pretty low!) $4.20 $88.20
Add GST $8.82 $97.02
Total Retail Price $97.02


Ooops! So our 50/20 unlimited service on the NBN, with a very low profit, comes out with a retail price of $97.02 – for a service TPG is selling for just $69.99

OK, so let’s work back from TPG’s retail price, remove all the known costs and see how much profit they are making (note TPG are likely to be using the bundles):-

Item Line cost Total
Starting (retail) price $69.99
Remove GST $6.36 $63.63
Remove NBN bundled 50/20 AVC $45.00 $18.63
Remove 2Mb of backhaul and IP transit (say $3.5/Mb) $7.00 $11.63
Remove billing, support, admin $10.00 $1.63
Margin after all costs removed $1.63


So that means even with TPG’s massively reduced costs (they own a lot of their own fibre), they are making less than $2/month (about 2.5%) profit on their 50/20 services. It is also worth noting that TPG’s service includes unlimited telephone calls. Few people use home phones these days, but it likely this will cost something. So it is even possible the TPG are really making no money at all or running at a slight loss.

Yes, all the above is back-of-the-envelope stuff, but it demonstrates the fundamental problem. This is why the small providers are going bust! It’s a bit of a case of the Emperor’s New Clothes here. New providers startup thinking that everyone else must be making money and they just need to get more efficient and increase their scale and all will be well. However, as they are finding, it’s not.

It is worth noting that NBN are hoping to increase their revenue (“ARPU”) from the current $44 per user to $52 per user. They may not be raising their prices, but they are expecting RSPs to hand over more in higher connection or bandwidth charges. The point is, NBN are unlikely to be reducing their prices here.

The only way to survive is to not play this price game. At Launtel, we found we had to connect directly to the NBN to improve our ability to provide a good service, be innovative and flexible in the type of service we provide that allows us to charge a little more. For example there is no way we could offer our daily pricing system through an aggregator. We knew that if we were competing on price, we would lose the game.

At some point something has to give here. Either NBN has to lower their prices (very unlikely) or the telco industry has to start raising their prices. If nothing happens the NBN market will continue to be dominated by the same big players whose only interest is to spend as little money as possible in providing the most minimal service they can get away with.

Damian Ivereigh