Occasionally a report comes along which should give people a whole new way of looking at a public policy debate. A new report on universal high-speed broadband (UHSB) via fibre-to-the-premises (FTTP), titled “Superfast: Is It Really Worth a Subsidy?”, does just that. Written by development economist Charles Kenny and his brother Robert Kenny, a British telecommunications consultant, it makes five points which are too frequently overlooked.
First, the macro-level studies which seek to show that UHSB via FTTP* will bring enormous economic benefits are mostly really ropey. This has been true for a decade and is getting worse. People who understand telecoms mostly don’t know how to do rigorous cost-benefit studies, or (like McKinsey & Co) don’t care.
Second, the micro-level specific benefits claimed for UHSB via FTTP in areas such as education, health and power management are mostly overstated.
Thirdly, many of the supposed benefits of UHSB via FTTP – macro- or micro-level – should be realised with the broadband we already have. It’s amazing how often the debate overlooks this. The report makes particular play of the way in which estimates of the bandwidth needs of electricity “smart grids” have been recklessly overstated. The same is true in other areas, including health. To make the case for UHSB via FTTP, proponents need to show what it can do that currently available broadband cannot.
Fourthly, UHSB over FTTP is frequently claimed as the solution to our toughest problems, the intractable ones with complex social roots. For example, it is supposed to transform health care, while health IT experts battle away vainly to get the industry to digitise its existing workflows (a project that really does look like it would have big payoffs). It is supposed educate our kids and cut our power consumption. At this point, UHSB over FTTP starts to look a little like a developed-world cargo-cult.
Fifthly – and perhaps most importantly – we should be able to see high-speed broadband (HSB) via FTTP at work today. The world has had HSB via FTTP for the better part of a decade in places like Seoul and Tokyo. And Australia, like most other developed countries, has had HSB via FTTP for many years too – not everywhere, but certainly between CBDs, many inner-city areas, and the universities. The notion that we can only imagine the future is only a half-truth. As William Gibson wrote long ago: “the future is here; it is just unevenly distributed”. So the incremental benefits of extra bandwidth for technologies like videoconferencing should be showing up in communications between different parts of Seoul and Tokyo, and between the Australian capital-city offices of major businesses and professional services firms. Yet these benefits are hard to find – harder, indeed, than I expected ten years ago. This is somewhat surprising, but also instructive. (The Kenny brothers make too much of South Korea’s recent slow growth; the more telling observation is the surprising paucity of new uses for South Korea’s shiny new broadband infrastructure.)
People – especially people poorly-grounded in the history of technology – can be easily misled into thinking that if the improvements of the previous generation of technology were good, then the next generation must be just as desirable. In fact, as the Kenny brothers point out, improvement in technology does not necessarily keep on delivering pay-offs in every successive generation; think of the Concorde, once seen as the logical successor to the turboprop and the subsonic jetliner.
The evidence in the Kenny brothers’ report and elsewhere suggests that the widespread adoption of the Internet in the late 1990s was the first-order communications revolution, and always-on middleband and slowish broadband was a second-order revolution. Both occurred incrementally in a short period of time, without great government command or subsidy; the Internet is well-suited to incremental improvement. UHSB via FTTP is a third-order issue.
I admire Stephen Conroy for his energy, his patriotism, and his courage. But his “national high-speed broadband plan” was conceived largely for the political purpose of aligning the Labor brand with “the future’, and costs far too much for what it delivers. It does not survive close scrutiny. That is why the government has been fighting to keep it away from the Productivity Commission, and why it should go there.
* Sorry, but I can’t think of a better way of putting it than UHSB via FTTP. “National broadband network” is a misleading name; we already have one, in the same sense that in 1996 we had a “national Internet network”. The question is whether we should pay more and restrict competition in order to deliver today’s gold-standard service to everyone.
Update: I had missed Ken Parish’s recent post and the ensuing discussion, which is one of the more detailed and well-informed you’ll find on the merits of the NBN.
Note that the Kenny brothers explicitly say that in many places there will be a commercial case for fibre right now, and that eventually everyone may need it. Their argument is that the case for government to subsidise or control universal HSB via FTTP is grossly overstated.
What you refer to as energy, patriotism and courage is something I think of as bluster, bullshit and bullheadedness.
Jacques, I don’t like a lot of Conroy’s current bluster either, but he has other qualities. And people who argue hard for something will often look bull-headed. (History only calls them that if they turn out to be wrong – think of Churchill.)
Note that my comment on Conroy’s patriotism is due to some exposure to him long ago. He came here from northern England in his youth, and regards the Australian suburbs as a sort of paradise, something I have always found peculiarly endearing. His patriotism is based on actual comparison data.
I’d rather this comment thread doesn’t become a discussion of Conroy’s persoanl qualities, though. It’s the policy that matters.
I don’t admire any policy of Conroy’s that I am aware of. I remain convinced that the NBN is needed to disguise any performance lag from his stupid filters.
I agree absolutely that it should go to the PC. The current price is beyond absurd. If you are ‘saving’ several billion on a project, there’s a problem.
I posted some time ago suggesting they consider a fibre to the node solution rather than the current Rolls Royce fibre to the home scheme.
There are quite a few significant current uses for fast broadband (let alone future ones) that AFAIK can’t be catered for adequately with currently available broadband or known enhancements thereof i.e. the existing copper landlines just aren’t up to it. For example CDU law school teaches 90% of its students via online virtual classrooms. Multi-screen HD video with minimal lag rates appears to need significantly more than the 12mbps maximum (with much lower averages) achievable by ADSL2+ or wireless broadband. We need consistent speeds of 24-30mbps to be able to teach professional skills online really effectively, in that something like 70% or more of human communication is non-verbal. It may well be that such functions have not migrated online to date because those sorts of speeds just aren’t available. We’re actually teaching skills online (the first university in the world to do so) but it’s only barely satisfactory with current speeds. I’m sure there would be numerous educational and other uses where a similar situation applies.
Although I argued that fibre to the node + newly available enhancements to ADSL for the “last mile” made those necessary higher speeds achievable, the redoubtable Possum Comitatus argued (quite persuasively I thought) that Telstra’s existing landline infrastructure was so degraded from years of underinvestment that this would not be a significantly cheaper solution. I’m wondering whether your analysis takes this into account or takes issue with it.
PS I absolutely agree that there should be a proper cost-benefit analysis done by the Productivity Commission. It should be possible for it to make some sort of reasonable assessment of broader social benefits and other externalities as well as blue sky/unknown future uses.
Further to your point and that of Possum.
There must be a good reason why Telstra fr some time have been putting in fibre not copper to its new customers considering the cost.
Further discussion by John Quiggin
They clearly have never encountered Telstra. They clearly have never encountered Telstra in Regional and Rural Australia.
Kenny Brothers Dismissed.
Completely correct, the NBN has made no plans to improve on what is currently available in terms of top speed. It does plan to make high speeds available to a broader selection of the community and spread the cost in proportion (and inevitably some of that cost will be carried by taxpayers).
The purpose is to close the “digital divide” by force if needs be… make sure no Australian gets something that other Australians don’t get. Productivity really never gets into the calculation… it’s a political calculation, not an economic one.
The main thing that bothers me is not the spending of tax money on fibers, it’s the slated destruction of an existing perfectly good copper network. If the opposition could do one thing only it would be to force the auction of existing copper on a suburb by suburb basis once Telstra has declared it will no longer maintain the copper. Destroying to copper is tantamount to pointlessly vandalizing a public asset. Sure NBN will wring hands and cry and moan that allowing competitive ISPs to buy local copper would spoil their wonderful business plan, but out the other side of their mouth they claim to have an unbeatably superior product. Let them demonstrate superiority in a fair fight.
If local ISPs could buy a suburb full of copper they would have customers flocking to them buying cheap 2M/1M DSL for about $25 per month (with a bundled VoIP phone and 10c untimed national calls). Telstra are well aware of this, NBN are well aware. That’s why they don’t want to let it happen — artificial scarcity. Stupid thing is that cheap 2M/1M copper links would do more for closing the digital divide (by honest free market competition) than anything government does sloshing around tax money.
As I pointed out last time we had this discussion. Those speeds ARE available. Have you tried bonded DSL?
Yes, faster services do cost more than slower services. Faster cars also cost more than slower cars. Bigger TV’s cost more than smaller TV’s. Lawyers seem to cost more than secretaries for that matter.