NBN and the election

Lots of commentary around Mike Quigley's Charles Todd oration yesterday. Nice to see that both Comms Day and itWire managed to record the fact there is one wireless provider that exceeds a 50G per month "quota" - with an unlimited plan.

Bevan Slattery has described it as "This is one of the most misleading, self-serving, factually incorrect speeches I have ever heard". In his Comms Day piece he takes up a number of points;
1. The consequence of Quigley's comment that the NBN will not make a commercial return
2. That Quigley's examples of AT&T and Verizon simultaneously investing in fibre and wireless are flawed
3. That the NBN has no understanding of wireless
4. That the NBN can't be good for competition

There is as always merit in each of these. I should note though that Bevan's position as Charles Todd Medallist commenting on the Charles Todd oration is not quite what it seems as the former is conferred by ATUG while the latter is to the TSA (now part of the ACS). It is possibly more reflective of our paucity of ICT heroes and/or the few people who read the res of the history.

But let's turn to the comments.

1. Quigley says the NBN will not make a "commercial return" and so Slattery says this reflects a breach of the original policy, a breach of the provisions for GBEs and a reason the investment should not be considered an "asset". In reverse order, an asset is anything with a future value including a cashflow from sale of services. The expectation of GBEs to generate commercial returns is so that they don't undercut competition where it can occur. Finally, the question of a "commercial return" is just too rubbery. If I computed the IRR of the NBN over its 50 year life I would get a "commercial" return. f I computed it over 5 years I wouldn't. What makes the investment not commercial is the time to get a return, not the NPV of its cash flows over time.

2. The AT&T and Verizon examples were ill-advised for exactly the reasons Slattery gives, these are vertically integrated behemoths who are being allowed to use fibre to crush the so-called CLECs. They have halted their fibre because I suspect they've already chilled the competitive investment by what in the trade they call a "credible threat". Recession in the US would also help. But nothing Slattery notes advances the proposition that it is wireless we are all waiting for.

3. The NBN's understanding of wireless I suspect is a lot better than some give it credit for. Quoting the dramatic increases in the peak speeds that are being bandied about for LTE is not the same as providing a lot of high speed links. More importantly, radio resource should in the longer term be focussed on use where it delivers its greatest value - in mobility. The ongoing process of managing spectrum by dividing it into little bits for different operators is also unsustainably inefficient - there is a lot of spare resource around because the networks are separate. Slattery and Corner are both right that there is still a lot of microwave backhaul used, but it isn't ideal in the long term. Long term vision would be fibre everywhere with one wireless network on top optimised for different levels of speed of the mobility.

4. Slattery continues to imagine that the DSL world is somehow competitive without really acknowledging that it is based on the same structure as a post NBN world. There is a monopoly provider of copper cable and a whole lot of firms trying to compete by using it.

Finally Slattery describes a "horror show" of the ultimate collapse of NBN Co with the only company big enough to bail it out being Telstra. But what is Telstra? It is the consequence of the investments originally made by the Commonwealth in the PMG. From 1959 on the PMG was entirely self-funding - it did borrow fro the Commonwealth but got no actual budget allocation. On its creation Telecom Australia took on the entire $4.5B debt to that time (1975). Over the next 13 years Telecom paid interest o the debt and repaid $1.5B in capital. On corporatisation in 1989 the $3B debt was converted to equity and Telecom continued to pay dividends. Before privatisation Telecom made a "special dividend" payment to the Commonwealth of $3B. That is the asset the Howard Government sold for something like $80B had a carrying value to the Commonwealth of precisely ZERO.

Finally Stuart Corner's excellent piece in itWire drew out the distinction between Moore's Law and Cooper's Law. he former says that processing power doubles every eighteen months, the latter that the carrying capacity of spectrum doubles every 2.5 years. The problem is when you add what I immodestly call "Havyatt's Law" which is written in far more arcane language as "What Intel giveth, Microsoft taketh away". The increasing in processing power is matched by a linear increase in bandwidth required. By definition, wireless can't keep up - unless you increase the density of transmitters.

Finally "graphic of the year" goes to The Oz when reporting the coalition policy.