Internode Blog

The NBN and structural separation – not just Telstra?

Saturday, May 8th, 2010 by

(This opinion piece is based on a post that was first published on the Business Spectator web site)

Its an interesting week for the National Broadband Network, with the release of the novel-length report into its implementation that the Australian Government commissioned and then sat on for two months before letting the rest of us see it.

What matters isn’t so much this report, it is the government policy that gets determined as a result of it. That policy, in turn, is a precursor to any real action (and in this context, the work happening to date in Tasmania is simply the skirmish being conducted to limber up for the real battle).

And it will be a battle. A battle against a large vertically integrated incumbent, in Telstra, plus at least one challenger who is also undertaking rapid vertical integration: TPG.

What those companies do about the potential for the government to force equitable wholesale access or demand penalties if they undercut NBN access pricing will be… interesting.

Telstra have spent the last decade demonstrating just how well it is possible to deflect, defer, avoid, and obfuscate ‘level playing field wholesale access’, as a vertically integrated company. The history of court battles in the past and the present chronicle their market power in this respect.

So much so that its quite obvious that the only way to get Telstra to play nicely in the NBN context is via structural separation.

Fail to do that, and the whole house of playing cards will fall, due to Telstra aggressively white-anting the process, simply because as a vertically integrated player, white-anting is more cost effective than playing nice.

But if it gets structurally separated (and really separated, not just a simulation), the two halves will modify their behaviour to become pro-competitive in their respective realms. This is because its the vertical integration that generates the anti-competition outcomes, and each half on its own just won’t act the same way as the combined entity does.

Now, lets play a bit of a thought experiment about vertical integration in the Australian telecommunication market, and see where it leads…

Rumours in the media this week say that TPG is on the verge of buying AAPT.

AAPT own two important assets and one important access right (other than a rapidly dwindling retail broadband customer base).

Lets see what that those assets would mean for TPG if such a transaction went ahead:

1) AAPT own some dedicated optical fibres on major intercapital paths (these were obtained from Optus decades ago).

This is the missing piece that connects the TPG/PIPE international fibre optic cable, across state lines, into the intra-state PIPE optical fibre network that hooks into most of the competitive DSLAM sites in Australia

TPG would then own fibre end to end from Guam all the way through to the DSLAM in your suburb, and all points in between.

Let that sink in for a minute.

2) They’d own a major shareholding in iiNet, via AAPT’s existing major shareholding in iiNet.

Let that sink in too.

3) They would have access to the zero cost ‘Gang of Four’ peering agreement that lets them exchange traffic with the BigPond and Optus national networks at (effectively) no cost, while all other market players are required to pay significant sums for that access.

(The ‘Gang of Four’ is the Internet backbone peering cartel that some claim no longer exists, but that my seven figure annual costs paid to access that cartel says otherwise. Access on a tenable basis to peering links is vital to high performance Internet access in an NBN world).

Next, imagine that TPG then made an offer that Amcom (who own the other major stake in iiNet) decide to accept.

Suddenly TPG owns an effective controlling interest in iiNet, and it completes the vertical integration process upward into multiple sectors of retail, as deeply as it has completed it downward, into multiple sectors of fibre asset ownership.

That combined entity would then be much larger in terms of customer count in ADSL broadband than Optus, along with holding that amazing vertical integration into its own metro, national, and international fibre networks.

All of this leads to interesting questions about just how vertically integrated a company has to be in order to be considered in the same basket of market power as Telstra, in the fixed line broadband market.

If such a vertically integrated TPG existed, just for instance… would the government then need to structurally separate them as well as Telstra, before building the NBN?

To avoid unfairly singling out Telstra for that treatment, perhaps it should do so with a (future, vertically integrated) TPG as well… but of course I’d hardly expect that vertically integrated TPG to take that notion lying down!!

We continue to live in interesting times.

(P.S. All of this is simply an exercise in speculation… nothing more. The author owns no shares in any of these companies. He is merely an interested observer).

Post tags: , , , ,