The Telecommunications Revolution

The outcome of the broadband network was announced today and I must say that I almost fell off my chair with delight. If I am reading the news correctly, the government is going to do it themselves with a fibre to the home network that will put Australia right up there with the world leaders in this space. It was only in November last year that I proposed this very same policy in AFR opinion piece (reproduced over the fold). The benefits include (a) avoiding the regulatory morass that is Telstra; (b) providing not only competition in broadband but also competition across the whole of telecommunications; and (c) a great way of providing a fiscal stimulus in anticipation of the coming recession.

All I can say is wow.

[Update: Looks like the NBN Expert Panel were on top of the key issues in January]

Plan B is 21st Century Communications

Joshua Gans, Australian Financial Review, 24th November 2008

It is obvious to everyone that, despite the best of intentions, the National Broadband Network (NBN) tender process is not going to plan. $4.7 billion being put on the table has sparked only limited interest and even that is fragile. Telstra is driving a hard public bargain on regulation and structural separation and may not participate. This is hardly a surprise as they are in a position to roll out their own network in the most profitable areas of the country and keep those profits for themselves. Other bidders, also not surprisingly, want to stop that from happening. It is a mess.

But the tender should not be the only game in town. If the government receives poor bids for the NBN, it can and should reject them all. There are alternatives that would fulfill on Labor’s election promise and maybe, in these troubled economic times, be a much better deal for the country.

Now I have been skeptic as to whether the proposed fibre to the node network covering 98 percent of the country with 12Mbps or greater download speeds is worth the government dollars. In my mind, it is a proposal stuck in the middle. It commits us to mediocrity (12Mbps will barely keep us at the average in the OECD and the price to consumers will still be high) and does not address the main economic benefit of government involvement: securing a competitive telecommunications (not just broadband) industry.

So let’s think big. What if the government rejected all bids as inadequate and embarked on building its own NBN? And not just one with mid-range bandwidth but instead a ‘world best’ target for a change: 100Mbps or even 1Gbps speeds. This would likely require fibre to the home in most places and, in others, some serious mobile technology. It would probably cost in excess of $15 billion but these costs would be spread over a number of years. An NBN could not be built in a day.

That is a big price tag but the advantages are many-fold. First, the government would control the price to users and could target low income households more effectively as well as providing services that were open if used for education and health (saving costs elsewhere). Second, the government itself would become a strong telecommunications provider in competition with Telstra. That will benefit homes and businesses everywhere. Third, that competition would allow us to dismantle most telecommunications regulations in the country once the network was operating.

But it is fair to ask: why now? Well, the world has changed since Labor made its initial broadband promises. Specifically, the social cost of government expenditure has dropped considerably as the need for fiscal stimulus becomes apparent. And this stimulus should not just be throwing money at pensioners and consumers. It needs to be done in a way that creates jobs and builds productive capacity. So when telcos inevitably howl over this proposed government investment in terms of lost jobs, the retort is straightforward: we will employ the people and the contractors. An army will move around the country laying cable and supporting the new broadband customers. Moreover, it will target key groups most vulnerable if private business should curtail its investing plans.

But wait, there’s more. With the global economy slowing, broadband investment elsewhere will slow. That will drive down the costs of purchasing equipment and optic fibre. It will also give us an opportunity to catch-up. Australia can emerge from recession with a 21st Century telecommunications industry, well-placed to take on the world. If we complement this with investments in capacity over the oceans, we can open up opportunities for video conferencing (saving on airplane emissions) and even a cloud computing facility located in cooler places such as Tasmania or perhaps, with cooperation, New Zealand.

By moving to Plan B, the government also has the opportunity to optimise its plans. While one possibility would be for it to have a state-owned corporation do the entire job, in reality, there are other ways to get a network built. For instance, the government could build or lease the backbone from existing providers. It could then use tax credits to encourage localised initiatives to build fibre or other high-speed options in the last mile. Last week, Google’s Derek Slater and Columbia University’s Tim Wu proposed just that. Their ‘homes with tails’ approach reconceptualises who owns the network. It need not be a provider but it can be users themselves. Consortia or local governments could provide the means by which neighbourhoods invested in fibre networks and then contracted out their maintenance and management. It would become a home owner’s asset in much the same way a new pool or a nice garden is. Consumers would recoup their investments in saved access fees as well as increased property prices.

For Slater and Wu, a real constraint would be how these local networks connected to the broader networks. But with a government-owned network, establishing Open Points of Presence could be done at the stroke of a legislative pen rather than through the uncertainty of years of negotiated regulation that would otherwise be required.

If broadband is of critical national importance as the government claims it is, the time for half-measures is over. A bold plan to leapfrog the competition and place Australia at the technological leading edge is both appropriate and timely given our macroeconomic circumstances. Fussing around as private enterprise tries to gain the best deal for their shareholders is wasting time and energy. Let it go Minister Conroy and move to a plan B that doesn’t just meet but exceeds your election promises.

Joshua Gans is a professor of economics at Melbourne Business School. His extensive writings on broadband can be found at

10 thoughts on “The Telecommunications Revolution”

  1. Two issues I see:
    1. The government network needs to include international undersea fibre optics to connect out to the world (about $4B), since Telstra and Optus current have a monopoly on them (think $0.15 per Mb).

    A fast Aussie network is great, but we want to see the world’s content without a private sector monopoly toll. Google is a likely co-investor in such as cable, as they have already invested in cables to Australia.

    2. A wireless urban build out using our home wifi networks, and smart handover would be useful. Think wimax, where the home router can reach out a kilometer or two, to blanket the urban area, with always on wireless, shared broadband. This is a government sanctioned home owned people net. The negative impact on the mobile phone business however could be profound.

    Also, the day is full of hope, but also peril. Creative destruction in play. There are project perils here – getting the right people, the right plan, the right project management. I think Tasmania is a great place to start. Let’s test the network design there, and a couple of other places, like Kalgoorlie, and Canberra, then roll it out further.

    A high level of project transparency is also heighly recommended. Let PeopleNet commence.


  2. I agree with the above post. A 2 order of magnitude increase in continental network bandwidth requires AT LEAST a 1 order of magnitude increase in the bandwidth of  intercontinental cable to the US. Currently the Southern Cross Cable is maxing out at 860Gbit/s. So 10 Tbit/s would be appropriate (that is 300,000 ppl streaming blur-ray movies)  . This is also the approximate current capacity of the total transatlantic lines.

    Furthermore if Australia wins the bid for the Square Kilometer Array (SKA) Radio Telescope to be built by 2020, there will be petabytes if not exabytes of data being generated somewhere in the Outback. This will need to makes its way to research institution around the world, and we don’t want it to clog up the P2P lines :p


  3. I’m a big fan of broadband but will this really generate returns to the economy greater than the $43bn cost? 

    As someone on Whirlpool observed:
    This looks like one third “meeting election commitment”, one third “economic stimulus” and one third “up yours” to telstra.”

    FTTH/P is a good for the reason you mentioned above.

    Richard/Jan Re: International capacity… I think that’s covered.  Telstra has built it’s own link (separate to Reach) and there is the PIPE cable.  International capacity seems to be coming along ok IMHO.


  4. Given the important point from the paper referenced;

    [FTTN is] unlikely to provide an efficient upgrade path to fibre-to-the-premises (FTTP), because of the high costs of equipment associated with rolling out
    a FTTN network that would not be required for a FTTP network (i.e. FTTN is not a pre-requisite for the provision of FTTP);

    I concede that this scheme would appear completely superior to my ideas. However one problem remains: the incentive to optimise the technology in the ‘tail’ coming from the home to the node (there will still be a node, unless you run a fat bundle of optical fibre direct to the exchange).

    In other words, can a new monopoly in fast broadband (Telstra won’t bother to compete, and own all the ducts anyway) avoid the problems of government-owned (or any monopolistic) infrastructure company servicing the public in terms of poor efficiency in service time and choice of hardware? Can we get some competition in the tails somehow?

    Perhaps it can come from the companies leasing the bandwidth and reselling it… perhaps they will champion the consumers they service? Or is the choice of technology a lay-down-misere as it was with twisted pair back in the day? Just how fast does optical fibre technology improve, and is this sufficient reason to introduce a node back into the mix so as upgrades are easier to effect.

    Of course, just like everyone I want to believe that the government will get it right – but we must be cautious and far sighted…


  5. invig

    Joshua will probably correct me here but wouldn’t the private aspect of the proposal address the perceived public inefficiencies point, ie, the company will need to be pretty efficient and reasonably profitable to get the 49% private investment sought?


  6. I like the way the new plan neatly gets around the rural lobby’s demand for a fibre-based solution to delivering broadband in rural areas.  Under the original proposal, FTTN was supposed to reach 98% of the population.  No one can seriously expect FTTP to be built in sparsely populated areas.  The new plan appropriately confines FTTP to 90% of the population and relies on wireless technology to deliver broadband to the rest of the country.


  7. The government can fund the broadband network and give the economy an immediate stimulus without going into debt.

    To do this the government issues every citizen in Australia with shares in the new Company with a face value of $2500. Anyone who wants the shares registers to obtain them. Many people will keep them and many people will sell them for whatever the market will pay them. Probably most will sell. This will create an immediate stimulus to the economy. The company will have a large amount of shares and when it needs some funds it will convert some of the shares to cash by asking the Reserve Bank to issue it with zero interest new money. This increases the money supply but only as assets are created to back the money. Because the money in the company will be spent on producing an earning asset this will not be inflationary. Thus the government can at one stroke solve the ownership problem of the new company, not go into debt, stimulate the economy, and solve the broadband infrastructure.

    This is a variation on increasing the money supply as explained in 

    The whole idea of the broadband network is to provide infrastructure at a reasonable price for all Australians and if you remove the capital cost then the last mile becomes very cheap. This means we can have broadband at very low prices which will increase productivity (and taxes) because the price of communications will drop. 

    If the last mile is already in place, such as in Canberra with Transact, then those assets can be purchased from TransACT with shares in the new broadband company. Similarly Telstra in those places where fibre is to the home can be given shares of equivalent value. This would accelerate the introduction of broadband through existing Telstra infrastructure as they would soon put in the 30 meters of fibre from a box along my street to my home and increase my broadband speed by 100 times. 

    As long as the broadband company ONLY owns and builds infrastructure and does not get into selling what is sent along the wires then this will give us the worlds cheapest communications infrastructure with all the economic benefits that will come from low priced communications. Giving shares to both the general population and to those whose assets can be used as part of the network is the fairest way of dividing up the ownership of the new asset created by through increasing the money supply.

    Politically it should be an easy sell.


  8. Dave55,

    It would be nice if that were true, but companies are not perfect market-oriented entities. They are full of people who like to be friends with those who can  offer them inducements in the form of RBA jobs, directorships in a new company or just plain bribes – whether individually or for their own company.

    With $43b being thrown around, I don’t think we can trust the motivations of anyone point blank.


Comments are closed.