The NBN, Joshua Gans and right-on industry policy

A while ago Paul Montgomery, whom I didn’t know, tweeted that he had wanted to set up a blog of the radical centre.  His tweet was about his crestfallen discovery that we beat him to it. Anyway, my handle @nichlasgruen was in this tweet so I saw it and suggested that Paul submit a guest post – which he did. It’s below.

One of the blog series nominated in the comments of the Best Blog Posts of 2010 competition, run by Club Troppo and On Line Opinion, is the series on the National Broadband Network by Joshua Gans at his Core Economics blog. Paul Frijters’ nomination went like so:

1. The series of blogs (eg. here, here, here, and here) by Joshua Gans on the National Broadband Network. He writes about his area of expertise, i.e. monopolies and government regulation, it’s a big national issue, and he mostly gets it right (I think).

“Right” is a matter of opinion, especially to those who would prefer an analysis to go beyond mere economic issues, as I would. The left in general has been fully behind the NBN from the start, with the policy being released by a card carrying member of the ALP Right in Steven Conroy in the name of microeconomic reform, but also fulfilling a lot of leftist ideals for government intervention to construct public institutions. The ALP Left seemed quite happy for the Right of its own party to take up the cudgel of the “vision thing”, while the mostly inner-city Greens could only applaud the extension of a key part of their own lifestyle to the masses, and the regional Independents put their provincialism ahead of their principles to swallow the Government’s promises of early roll-outs to the bush. Thus most of the inquiry into the NBN as a policy issue has come from the right, for better or worse.

As the major policy item in a federal election this year which was otherwise bereft of serious discussion, the Coalition did not do a very good job of prosecuting their anti-NBN case in the eyes of the electorate, if the pre-election opinion polls and the post-election exit polls on the issue were anything to go by. The Coalition itself was not the best antagonist to argue against the NBN, as Malcolm Turnbull is its only parliamentarian qualified to speak on the technicalities and his performance was weighed down by his obvious political baggage. The Australian, a newspaper that prides itself on the power and influence of its frequent campaigns, could not mount a sustained attack on the NBN despite having many of the finest IT journalists in the country in its ranks. The media has let the public and/or the conservative side of politics down by not scrutinising the NBN enough, allowing some very rubbery figures to go by mostly unchallenged.

There are some elements of Gans’ analysis with which I fully agree. No, the NBN itself will probably not make a profit – or if it will, it will require such draconian anti-competitive measures to ensure it hits take-up targets so as to even alarm its cheerleaders on the left. The low take-up in Tasmania, if extrapolated to the rest of the country, can not justify the investment prima facie. Additionally, because so much of the investment is private sector and the government has already committed unofficially to give investors a better rate than long-term bonds, all of the losses will be eaten by the public, disproportionately to their capital contribution.

Yes, there is a good chance that the NBN will pay for itself in the wider scheme of things, which is the only real justification for making the NBN a public project. This is the Tiger Woods argument. Just as the Victorian government paid a $3 million fee to Woods’ appearances at the Masters golf tournament for the past two years, which produced an estimated $20 million the first time around, similarly, the federal government may be throwing away money on the NBN itself, but it will come back in the form of extra tax revenue from all the economic activity that the NBN will engender. Those sums could even add up without considering the so-called “social return”. That’s all very well, but it’s not the argument that we have been having in the national media, as they are still stuck on black or red ink for NBN Co. As has been pointed out before many times, this gamble is based on some airy-fairy projections, but then again so was the Coalition’s policy. It was not a great choice for voters to have to make, but they seemed to like the blue sky of the Government’s rhetoric rather than trusting broadband to be delivered wirelessly over the sky itself.

If we did have the “real” argument, it might actually do some good. I think the NBN as a project is not enough on its own. Building it and hoping that new local IT industries will pop up overnight like remora on a whale shark is rather poor policy. There needs to be a concurrent effort on encouraging these new industries with tax breaks for investment and support for education and nurturing of young Australian talent at secondary and tertiary levels. Computer science education in this country has been a basket case since the beginning. You’re almost better off teaching yourself and not wasting your time at university if you want to be an agile entrepreneur.

There has been some talk during and after the election about these ethereal new industries that will somehow conspire to deliver a return, financial and/or social, for the massive investment being made in the NBN. However, neither side has made any specific policy to actually make this happen. Why is that? Is it because the dollar values of the network itself are so humongous that it is anathema for politicians to even think about adding on extra investment in human capital on top of the cash ploughed into the ground? Is it thought to be enough effort on the supply side to just build the network, so that any further sweat of the brow should be spent on building demand?

I am surprised that the likes of Conroy and Gillard have not latched onto the media-friendly aspects of supporting an industry policy to underpin the project, if only at a superficial level. Even if the ALP was going to be highly cynical and only pay lip service to the ideal, it seems like low-hanging fruit to me to start talking even now about how the NBN-delivered future will be full of promise for new industries built on its back, like it was a million-polygonal pixellated merino. Perhaps leading up to the next federal election, the protagonists of the NBN will conduct a series of whistle stops for cutting ribbons of new innovation centres, shaking hands with the likes of serial tech midwife Mick Liubinskas or angel investor Tony Faure while announcing support for early-stage technology startups, and delivering keynote speeches exhorting the yoof of today to become the captains of tomorrow’s industries. The ALP has shown little sign to date that it is going to follow this path, despite the NBN being by far its most popular policy of the 2010 campaign.

While Gans worries about the immediate issue of subversion of the Trade Practices Act in setting up the NBN, I am more concerned about the lack of competition after the network is built in the secondary and tertiary industries that the project will require to be successful. Is delivery of the crucial IT applications like e-health, B2B and government services going to be left to the giant foreign-owned companies like IBM, Deloitte and HP? Given its history, Telstra’s ambition must surely be to create itself new monopolies in the NBN app market, but what would the Government’s response to that be? Instead of these vital questions, we’re still left debating copper versus fibre and watching the same old COAG infighting.

The major economic booms in Australia have been built on the efforts of the average worker: it was cockies riding on sheep’s backs, it was every man and his dog in the gold rush, and currently anyone can get a job working in the mines. I’m not sure I agree with Gans’ conclusion that the NBN is a subsidy for the rich, as he’s talking about a different set of rich than I am, but if the NBN is going to usher forth a new future where we transition smoothly from the minerals boom to a longer and more sustainable future based on being the clever country, then the Government has to do more than merely rolling out the cables to ensure it happens. It must support the plastic and glass with some human investment as well.

Paul Montgomery is a former IT journalist and a former ISP executive, and is currently founder and editor of FanFooty, a Web site about Australian rules fantasy football.

This entry was posted in Economics and public policy, IT and Internet. Bookmark the permalink.
Notify of

Newest Most Voted
Inline Feedbacks
View all comments
Joshua Gans
13 years ago

Paul you really need to read my posts more carefully. I am not the launching point for disagreement. I have been calling for a “real debate” away from commercial criteria for months.

On the issue of it being a boon for the rich. If the NBN charges monopoly rates, then the poor won’t get any benefit from it. If the NBN is geared towards video downloads, then there won’t be any of the spillovers hoped for.

What I have wanted was for basic broadband to be free. I know that makes me some radical left-winger in the eyes of people who want to paint economic issues in political terms but so be it. I’ll take that hit. As an economist, I am in favour of spending public money for public rather than private benefit. The politics have sadly driven us to a place where the NBN is a handout to subsidise private consumption activities. Tragic.

Paul Montgomery
13 years ago

As you would have seen Joshua, I did list a number of points on which we had major agreement. The reason I used your blog series as starting point was as much because it was a quality analysis to inform the discussion than any serious philosophical difference with you.

The NBN is not going to charge monopoly rates, in the sense that we have come to know from Telstra. In my experience reporting on these things and in the ISP industry, Telstra has always charged too high in an effort to dampen demand to a level they are comfortable in provisioning for, esepcially as far as high-end Internet applications are concerned. The Telstra era of the Internet in Australia was one of forced scarcity, whereas the NBN era will be one of abundance. There will be too much pressure on the Government to get serious levels of take-up with the NBN, which will be very price-sensitive. If anything, the tension is on the other side: that they will not charge enough to make the return worth the investment.

The NBN will be geared towards high-bandwidth applications, including big file downloads. That’s part of the deal. Are you referring there to the network being wasted on BitTorrent? That’s going to be a part of it, for sure, but all the more reason to encourage new, locally-produced applications to take up all that bandwidth for something more productive than siphoning new episodes of The Walking Dead.

Free basic broadband is a rather radical proposal, yes, but TNSTAAFL. Does it matter if there’s 80%+ take-up whether the taxpaying public pay for it through bills that go through NBN Co, or through their tax returns? They’re going to pay for it either way, and if the Government has anything to do with it, the maximum amount of voters possible are going to use it.


[…] post: Club Troppo » The NBN, Joshua Gans and right-on industry policy Share and […]

Beau Fabry
Beau Fabry
13 years ago

Computer science education in this country has been a basket case since the beginning. You’re almost better off teaching yourself and not wasting your time at university if you want to be an agile entrepreneur.

Well now. If you want to be an “agile entrepeneur” there’s a fair chance you’re correct. But if you want to get an education in computer science I’d still start with the University degree.

Paul Montgomery
13 years ago

Beau, I guess that leads to a separate discussion about what CS courses should be teaching. I have heard time and time again from people who went through such courses in Australia (and admittedly OS too) that the industry moves much faster than CS course syllabi, such that the languages that are taught in CS courses are often three or four waves removed from the latest and most important ones that programmers need to know to get the most interesting jobs – or to create their own.

I suppose the question is whether CS course designers should have to make a choice between preparing you for life as a salariman in a large, ossified corporation doing things as they were done 10+ years ago, or following the hottest trends in the industry in an effort to train kids with enough current knowledge to make them valuable to startups and other innovative companies. The latter is far harder, of course, and is often in the too-hard basket for universities.

13 years ago

You guys are hedging around it — the NBN is not commercially viable unless it can achieve a monopoly position on “last mile” fixed-line communications in Australia. The deal with Telstra is essentially a situation where Telstra is getting paid not to compete, and worse than that, all other ISPs are being forced into a position where they also are going to be unable to compete (because their access to copper will be taken away).

If the newly installed fiber infrastructure was forced to compete head-to-head with the incumbent copper infrastructure then I absolutely guarantee copper would hold onto at least 70% of the fixed-line market, purely by offering a cheaper price. Copper needs only pay for maintenance of what already exists, fiber needs to pay for rollout of new technology, and also maintenance of that technology. The vast bulk of the low-end communications market is highly price sensitive, people are perfectly happy with speeds around a few megabit, they just want it at the lowest monthly price.

In any other industry, for a private player to be paid not to compete would be absolutely unacceptable, but somehow in the rush to believe the “something for nothing” fairy, many people are carefully ignoring this fact.

Beyond this, there seems to be an expectation that a new crop of high bandwidth applications are going to justify both the cost of new fiber rollout and the bloody-minded destruction of perfectly good existing copper infrastructure. The problem is that high bandwidth communication is already available right now to people who want to pay for it. Firstly, you can have fiber (at a price) pretty much anywhere in metro areas right now with minimal fuss. Secondly, you can have a range of bonded copper options (allowing you to run multiple copper pairs in parallel to increase bandwidth). Thirdly, large institutions including nearly all our universities and a good fraction of businesses already have extensive ethernet infrastructure on campus, plus largish links to the wider internet “cloud”. Fourthly, other countries are ahead of Australia in this regard and they have already had years to think about what to do with their high bandwidth infrastructure.

Don’t you think that if a crop of killer applications were on the horizon, then they would already be showing up in the existing system? At least somewhere we would have something to point to and say, “That, we want that”.

Think about the whole e-health scenario… are you seriously going to sit home with a webcam pushing your own tongue down with a paddle pop stick? No, you are going to go to some local clinic which will have a handful of broadly trained medical staff and the only time they need high bandwidth communications is when the general staff need to call in a remote specialist. The only data transfer will be from specific clinics to existing hospitals… so fiber to the home isn’t even part of the whole e-health pipe dream. By all means run fiber into medical centers, but don’t do this by stealth under the guise of a fuzzy-return National Broadband Network, do it because there’s a specific return on fiber into medical centers.

Paul Montgomery
13 years ago

Tel, you have more faith in the quality and lifespan of the copper network than others, notably Possum Comitatus:

The copper network is on its last legs. It is a decrepit, broken thing attempting to provide something it was never designed for, whose performance has already peaked and which will become increasingly degraded and prohibitively expensive to maintain over time.

Fibre has always been the long-term answer to replace copper. It was just a matter of when, and by whom. The Government’s decision to hold the TPA in abeyance to make the NBN sums work is only surprising in that the Government was bold enough to do what needed to be done, in my opinion, as it has a decent argument behind it to ensure the investment makes sense right now. I mean, sure, you could wait until the copper is completely decrepit – some would argue it’s almost there as it is.

We don’t know what applications will succeed on the NBN. It’s useless and counterproductive to pick winners or losers at this stage, even in areas crying out for it like the health sector where there is a lot of wasted effort and resources at the moment.

Finally, this is not an area in which we should take refuge in the bad old Australian habit of the cultural cringe. That’s part of what I’m raging at: the lack of faith in Australians to create innovative businesses in new markets.

Paul Montgomery
13 years ago

Road-based analogies are often, in my experience, host to the most egregious examples of the form. Then again, I have suffered through a lot of “information superhighway” analogies in my time.

If you want to use a road analogy, switching investment away from copper to the NBN would be like slashing government spending on unsealed dirt road maintenance, and more on maglev trains with a train stop on the doorsteps of 95% of the country. :)

Paul Montgomery
13 years ago

Competition in the DSL market is based on continuing investment by Telstra in maintenance of a rapidly deteriorating asset, for decreasing levels of performance. There is a law of diminishing returns in effect here. Sure, you could wait until X% of the copper is unusable to squeeze the last percentage point of ROI out of the remainder, and while that would satisfy some economic rationalists who seem to treat copper wires like they’re rare earth deposits in the Kimberley, that would be a terrible result for the public as they get less and less access to cheap high speed broadband.

Telecoms is a strange industry in that the normal rules of competition don’t apply in certain aspects: every customer has to be served, regardless of need or ability to pay. This causes weirdness like the government ignoring the TPA in the interests of competition, or the freakiness of the Universal Service Obligation. The government has often discouraged cherrypicking in telecoms, especially of the high-margin city sectors, as a lot of the large-scale dynamics of the market are based on cross-subsidisation: rich and poor, city and country. Leaving the POTS network open to compete with the NBN would allow the worst kind of cherrypicking, one which undermines the ability of providers to provide the same service to all customers.

Nicholas Gruen
13 years ago


Why make a decision on such technical matters at the centre. If it’s a deteriorating asset, why not let it deteriorate and if its owner can’t make it pay they’ll give it away. Why make a national decision to close it down when it evidently has a valuable use? Anyway, you’ve answered the question. You wan to do it to fund cross subsidies to people in the Kimberley – who won’t really get access to the NBN but will pick up a subsidy for some kind of wireless service. If that’s such a great idea, (and I doubt it is, at least in the Kimberly) why should it be funded from the network as opposed to general revenue?

Paul Montgomery
13 years ago

Nicholas, that would lead to the copper being maintained only in the CBD and near business parks to cherrypick the most lucrative clients, attacking the NBN’s best revenue sources and undermining a lot of the project’s business plan targets. How could the government propose funding the NBN in the first place without preventing such effortless hamstringing of their potential to return the public’s investment?

I am guessing you mean funding from “general revenue” to mean government subsidies? The more likely possibility (which you may be referencing) would be the USO model, whereby other carriers are taxed to fund the operations of the carriers who are providing the unprofitable remote services… which is a cross-subsidisation imposed on the network. There is enough margin in telecoms to cross-subsidise service provision without taking money out of government coffers, while still maintaining an excellent ROI across the industry. Regulation by governments of either flavour has always been to ensure that the responsibility of carrying that burden has been shared, one way or another. Shutting down the POTS is just another measure in a long history of such policies.

Nicholas Gruen
13 years ago


I think you may be a mercantilist. You’re trying to make ‘the project’ work. And a whole other network gets destroyed to make this new network work.

The idea of the network is to contribute to Australia’s economic and social wellbeing. If that’s the case and you want to subsidise the bush, why don’t you subsidise the bush from central revenue? Why destroy an asset that’s serving a need and disciplining prices into the bargain?

When was the last time we destroyed an economic asset to help some other asset generate a commercial return? Seriously – it must have been decades ago. Oh wait – it was closing down the cross-city capabilities of various Sydney roads so that the numbers for the cross city privately financed tunnel would look a bit better. That was a success;) Not for the people in the cars of course – but for the guys financing the tunnel. It improved the business case – think of how much worse it would have been without forcing traffic onto the tunnel.

Paul Montgomery
13 years ago

Using examples from recent NSW state politics is always going to make the government look bad! ;)

Supporting universal service via cross-subsidisation is not a sign of mercantilism, in my opinion. It is a policy that has been adopted by governments of many stripes around the world for many decades. Mercantilism seems more of an -ism of inter-country relations, whereas we’re talking here about intracountry policy. There may be another -ism to describe it, but you’re the political blogger so you tell me. Socialism, perhaps?

The word “destroy” is an emotive one in this context. It’s not as if Mike Quigley is going to drop the plunger to ignite sticks of dynamite to blow up a mountain of copper wires. I could equally use language like “turn off the life support” to denote the undead nature of the network, which has lived long past its use by date and is being animated, Frankenstein style, to do things it was never designed to do. More accurately, the POTS is going to be allowed to degrade past its natural point of usefulness on its own.

I guess what we’re arguing here is what the primary goal of industry regulation should be. If it’s to minimise prices, that may be achieved for the cherrypicked sectors of an industry with the POTS still switched on, but what would it do for the edges of the network which are not cross-subsidised? Those customers would get stiffed, relative to what they enjoy now. If it’s to maximise competition, that may be true for the cherrypicked sectors, but again, the edges will have zero competition and will pay more for the NBN which would have to raise prices to give its investors any hope of ROI.

Telecoms is an industry where there will never be perfect competition. Due to the way governments around the world introduced the technology in the first place, there is always a dominant formerly public incumbent which has polluted the market environment to such an extent that the government is always playing catch-up to prevent the more damaging excesses of monopolistic behaviour.

It would be nice to have a telecoms industry where the ideals of competition can be maintained to the standards that Joshua Gans and other economists would like. Unfortunately, we live in the real world, where solutions have to be jury-rigged under less than pristine circumstances.