The devil in the NBN's detail

Remember the ‘paperless office'? The term gained currency when desktop computers started popping up on desks around the the world and we thought everything would be read on-screen.

By the mid-1990s, many noted the irony that computers, and a new generation of printers, had filled offices with more paper than ever. 

So doubts began to grow. In May 1995, The Australian wrote: “The paperless office and cashless society have been much talked about but to date the reality has fallen short of the rhetoric.”

With the benefit of hindsight, however, we can see that the paperless office, and cashless society for that matter, were just taking longer to arrive that we'd hoped – both have been spurred on by smartphones and tablets in particular.

And so it is likely to be with broadband and the myriad applications we already use – or rather, that some Australians currently use.

Streamed video, which seemed a luxury five years ago, is easy. High definition streaming is easy for some. And streamed super-high-definition 4K TV is possible, for a few.

Broadband is about a lot more than watching movies, though the network congestion that movies cause means that video consumption habits correlate pretty well to the kind of network we need access to.

And yet the network Australians have access to is falling further behind each year. As explained last week, both Coalition and Labor governments are culpable.

Both sides of politics want the voting public to take on board complex arguments as to why the whole mess is the fault of the other side.

That game would be terrific fun, were it not for the enormous opportunity cost to Australian consumers and, therefore, businesses.

One of the great myths of Australian broadband is that anyone who really needs it, can get it.

That is, if you are a corporation, a hospital, a government department, a university, or even a small business, it is relatively cheap to connect to an all-fibre network with virtually unlimited data-tranfer speeds.

That in itself is true, but it misses the point that the corporation or SME's customers, the hospital's out-patients, the government department's ‘stakeholders', or the university's students need a world-class network to easily connect with all these institutions.

The opportunity cost of not being able to connect effortlessly is immense. Capital cities are grinding to a halt while telecommuting – to take just one major productivity enhancing role for broadband – has been discussed for decades but hampered by sub-par technology.

Which brings us back to the nation's greatest ever infrastructure project, the NBN.

Labor promised something like the world's best all-fibre network, at huge expense, but delivered years, if not decades, into the future. Part of that was the ‘roll in' principle foisted upon the Gillard minority government by regional crossbench MPs.

But again, consumers don't care why their internet hasn't been upgraded – just that it hasn't been upgraded.

And so incoming communications minister Malcolm Turbull went to the 2013 election with a plan to provide cheaper, improved broadband, sooner and at less cost to the taxpayer.

The 93 per cent fibre network that began construction under former minister Stephen Conroy was popular with voters, but since not much of it was built before his government was turfed out, voters have mainly been sanguine about Turnbull's offering – a lower-spec network that should arrive much sooner.

Turbull argues that the 50 Mbps downloading he intends most Australians to have in the next few years is enough, pretty much forever.

That's because although data use is increasingly exponentially, compression technologies and new hardware solutions are also increasing pretty rapidly.

The Vertigan panel, which was instructed by Turnbull to map out the policy directions for his more thrify ‘multi-technology mix' (MTM) plan, thinks most people will not use more than about 15 Mbps by 2023. It concedes that extreme downloaders will need more – but hardly any would use more than three times that average 15 Mbps.

Those sorts of details are thrown around the in media, and this is half the problem in moving this appallingly handled area of public policy forwards.

Because whatever speed Turnbull thinks is enough, or whatever speeds are suggested by the geek army who sneer at 50 Mbps, speed is far from the central issue.

Put another way, if Turnbull could deliver even a reliable 20 Mbps to all those customers, patients, students and stakeholders (urgh, how I hate that term) in a short time, he would be feted as a historic reformer.

These issues have been clear for a long time. A good level of upload/download speed that was truly ubiquitous, delivered by a network that did not jam up at peak-data times, is what the debate should have been about all along (see: Time to untangle the NBN, December 2010).

While we don't have such a network, we pay an invisible price in the opportunity cost of not being able to tele-commute more, ‘visit' a health professional on line, reliably attend online tutorials with a university educator, or enjoy online shopping experiences that retailers will build ‘one day' when the network is up to it.

Why mention all this now? Because as Gary McLaren explained

http://www.businessspectator.com.au/article/2014/10/8/technology/breaking-nbn-co-needs-immediate-action

on Wednesday, the history of failed telecommunications policy as at another fork in the road.

The key to understanding it is to grasp the significant difference between the Conroy and Turnbull projects – and no, speed is not the main difference.

The two choices offered to the Australian people were these:

– Break Telstra's monopolistic power by paying it to shut down key assets (the ‘last mile' copper and HFC networks), buying access to its ducts and pits, and building a mostly fibre public monopoly against which it would either be futile or illegal to compete. Short version: build a public infrastructure monopoly likely to be less efficient than a private sector telco, but which provides complete certainty for the taxpayers who own it.

– Or, break Telstra's monopolistic power by buying its last-mile copper to connect to a fibre-to-the-node network, buying and using its HFC network, and using a range of other technologies as appropriate. Short version: build a cheaper public monopoly with a strong bias towards privatising it later, but with some risk that private monopolies might arise in future.

In the above, I'm intentionally leaving Optus out for the sake of simplicity – its assets are smaller and its bearing on the overall market failure far less.

The former option above, pushed by Labor, was expensive and quite likely to produce a flabby government owned enterprise similar to the old Telecom. However, it would have the advantage of removing a hugely complex layer of regulation that had been used to ensure a ‘free' market in which telcos could operate. Labor mumbled about privatising this monopoly at some point, but it was an afterthought.

The latter option, now being pursued by Turnbull, was reviewd by the Vertigan panel, which came to the conclusion that, since it did not ban some significant kinds of competition from other telcos, was probably best sold off as soon as possible. That would allow real infrastructure-based competition – the most common form of telco market around the world.

One of these two options must triumph, because only one or the other will deliver the investment needed to get Australian broadband moving again – one a government-mandated investment, and the other private.

The problem is, Turnbull thinks that choice can be delayed, so that the government once again faces the prospect of privatising a complex, multi-technology giant. Vertigan would prefer the various technologies to be sold off as soon as possible, while some Turnbull supporters are calling for NBN Co to at least be built in units easy to hive off - "competition ready" is the phrase.

Importantly, deciding not to focus on that issue now means, de facto, creating a hybrid monster that is neither an expensive, low-risk fibre network as Conroy proposed, nor a market of free players able to compete by normal market rules.

Conroy banned Telstra from biting off chunks of his super-network by legislating to prevent it (or others) marketing wireless solutions as an alternative to the expensive fibre network.

The problem is, the same threat challenges the Turnbull almost-public-monopoly. It is having chunks bitten off even as it tries to expand to fix the underperforming national network.

One submission to the Vertigan panel report, from Malaysian-based investment bank CIMB noted: “Developments in wireless technology are likely to expand the opportunities to bypass NBN Co in any case. NBN Co's effective footprint is likely to shrink however much the government tries to preserve it.”

Moreover, TPG is expanding its network after it found a loophole in the existing legislation that allows it to expand existing infrastructure by 1km. It has worked out that it can add up to 500,000 apartments in the inner cities by running fibre into apartment block basements.

The government is imposing special licence conditions on TPG to partially counter this move, but it still undermines the economics of NBN Co, and avoids paying the cross-subsidy for the bush that NBN-connected RSPs will pay.

So we are now building a whale, being torn apart by sharks. McLaren's argument is that we either stop the sharks undermining a giant public investment, or cut our whale into shark-sized pieces and release them into the water.

Central to this whole long, sorry tale is the role of regulation. The ACCC, which was given the role of containing Telstra's market power in 1997, is heavily criticised in some quarters for capricious or even perverse rulings.

Two examples: to create a ‘level' playing field between Telstra and access-seekers, up until 2010 it just kept cutting the cost of access, below cost, until the access-seekers seemed able to compete.

Some claim that this resulted in Optus re-selling copper-based internet to customers who had its HFC cable running past their doors – because re-selling cut-price ADSL was more profitable. That is perverse.

Further, the calculation method the ACCC used prior to 2010 did not consider Telstra's need to make a return on its investments – only what would allow the access-seekers in. So nobody invested anywhere near as much as was required to meet demand.

So where does that leave the Turnbull model? Unfortunately, in limbo.

Turnbull still needs to negotiate with Telstra to buy the copper for the FTTN network - Telstra has indicated that will essentially be the same figure Conroy negotiated for it to leave the copper lying idle in the ground.

He then oversees the building of a massive network that, as it is rolled out, is already facing competition from wireless and the TPG play.

Worst of all, Telstra, soon to be a retail-only player, will be the most cashed-up RSP in the country thanks to all those tax dollars it gets for selling its high-maintenance copper lines. Oh, and a dominant wireless player.

Put side-by-side, the Conroy plan is a simpler, more secure investment for taxpayers, whereas Turnbull's plan is more complex, more risky, but a lot cheaper.

The politics of the NBN from 2009 to 2013 was all about painting the NBN as a ‘Roll Royce' stuffed full of tax dollars, and Conroy as the devil driving it.

However, the major risks of ongoing market failure that threaten the Turnbull plan – if they prevent a combination of private and public broadband investment catching up with other developed nations – may leave voters longing for the ‘devil they know'.