Does Palo Alto's FttP rejection hurt the NBN?

Does Palo Alto's FttP rejection hurt the NBN?

Summary: It turns out that the city at the heart of Silicon Valley can't afford to roll out FttP to its own residents — but how does that affect the financial viability of our own FttP NBN?


News that the city of Palo Alto, California has all but rejected the idea of pervasive fibre to the premises (FttP) will be welcomed with open arms by opponents of the National Broadband Network (NBN) — but is it an indictment of the whole idea of an FttP NBN, or a validation of the unorthodox strategy being taken by NBN Co?

Turns out that the local government of Palo Alto — an affluent community in the heart of Silicon Valley that's ironically also the manufacturing base for NBN Co satellite builder Space Systems/Loral — believes that it's not going to be cost effective to expand its FttP network any further. Its decision apparently stems from an independent report that concluded, "A fully user-financed citywide fibre-to-the-premise system is not possible to achieve in Palo Alto. An opt-in FttP system can be built using a combination of upfront user fees and city financing, but there is very little probability of the debt incurred being repaid through operations."

Palo Alto's councillors have concluded that FttP is financially nonviable; does that kill the case for Labor's NBN? (City Hall image by Nimur, CC BY-SA 3.0)

Sound like a Liberal Party press release? It gets better. The report considers a best-case scenario in which homeowners pay US$1000 to get an FttP connection, and then pay US$75 monthly for their subscription. At a take-up rate of 20 per cent, the report concluded that such a system would lose several hundred thousand dollars per year. And that's just in one city.

Given his history of latching onto economic and overseas political reports to justify his opposition to Labor's FttP plan, I have no doubt that Malcolm Turnbull will rush to say that this report is further evidence that the current NBN is a half-baked, overdone and far-too-expensive venture. If they can't even justify FttP in the heart of the technology world's innovation capitol, he may argue, how can we justify rolling it out across an entire country?

Citing the report's financial modelling and monthly subscription prices that seem largely in line with what Australian NBN providers are paying — all magnified by a lack of Australian co-contributions, and the vastly larger scope of our project — such an argument might seem to be right. But there are several other variables that reflect just how different the Palo Alto experiment is from our own.

Like all major telecoms investments, FttP is desperately uneconomical when done in small quantities.

Consider, for example, the localised nature of Palo Alto's roll-out. That city's population is just 64,403 people, spread over about 26,500 households. This is hardly a sizable population — many Australian councils have four or more times as many people and households — and it's a disaster for an FttP roll-out, because the base costs of the optical equipment alone would skew the business case towards financial disaster.

Like all major telecoms investments, FttP is desperately uneconomical when done in small quantities. As Conroy pointed out two years ago, and continues to argue: the Tasmanian NBN would be nonviable by itself. Indeed, Mike Quigley made claims to a recent Senate Estimates hearing that the NBN would not be commercially attractive if sold off incomplete.

America's highly decentralised local-government layer tends towards disarray; it is only large-scale operators like Verizon that have the geographic spread to roll out FttP over a large area with something resembling cost effectiveness — yet, even their interest in unilaterally pushing FttP is limited, because they have extensive investment in existing copper and cable infrastructure.

Put simply, the US just does not need fibre; with cable infrastructure well-established in the US since the late 1970s and available to nearly every home, cable and local-telephone services are already in competitive lockstep for broadband customers. There really is no place for a third network topology. For them, FttP is a telecoms frisson for those Americans who want a third broadband option.

Much of Australia, by contrast, doesn't even have a second option. And just because the US does not generally need fibre, does not mean that we are automatically in the same boat. Despite what Turnbull would have us believe, Australia's cable infrastructure is a pale imitation of that overseas.

It is only Telstra's cable that even reaches most capital cities; Optus' network, the closure of which caused Turnbull and others to recently raise such a stink, is only available in limited parts of Sydney, Brisbane and Melbourne. Not even Optus sees value in maintaining that network long term if it can get NBN Co to do its dirty work.

In other words: HFC's limited scope has generated none of the competitive pressure that Palo Alto's fibre has faced from existing alternatives, and it would never do so, no matter how much a coalition government would try to prop it up with subsidies.

(You could, perhaps, reproduce some of the US market's competitive tension by selling existing Telstra, Optus and TransACT cable to a new third-party operator, but good luck finding a commercial operator that has any interest in that idea.)

In truth, Palo Alto's step away from FttP is not a condemnation of the technology or its viability, but rather a general reflection of the limited opportunity to introduce a completely new communications topography into a market that is already effectively serviced by alternatives. But Australia's market is not already effectively serviced by alternatives, and the existing infrastructure is on a doomsday countdown that not even Telstra wants to be a part of.

If FttP is therefore a logical next step, it then follows that the NBN must be accompanied by changes to the competitive model, so that the network can be assured of high take-up and cross-subsidised low prices. These changes may raise the hackles of pro-competition cheerleaders like Turnbull, but at some point the government must consider a compromise in the name of outcomes.

There will be times when consumers' best interests are served by managing competition, rather than promoting it at all costs.

As we've seen over the past 15 years, stubbornly sticking to the competition-at-all-costs line is unlikely to deliver the type of consumer benefit for which deregulation was originally designed. Competition compromises may seem unattractive, but there is a good case to be made that they are necessary to deliver the outcomes that Palo Alto could not; hence the ACCC's concession that it would serve little good to block the NBN Co/Optus HFC deal.

Perhaps the conflict comes from the ACCC's design itself: one of those Cs stands for "Consumer", and another for "Competition". But there will be times when consumers' best interests are served by managing competition, rather than promoting it at all costs.

It may not always uphold the Coalition's free market ideals, but — as Palo Alto has learned at great expense — a fundamentally objective and pragmatic approach to competition, and one that is not afraid to consider tweaking it as well as encouraging it, may prove to be the only way to chart a viable communications future.

What do you think? Does the Palo Alto decision taint our own FttP NBN project? Or is it just a reminder that we need to take overseas comparisons — and blind allegiance to competition — with a pinch of salt?

Topics: Broadband, CXO, Government AU, Telcos, Optus, NBN


Australia’s first-world economy relies on first-rate IT and telecommunications innovation. David Braue, an award-winning IT journalist and former Macworld editor, covers its challenges, successes and lessons learned as it uses ICT to assert its leadership in the developing Asia-Pacific region – and strengthen its reputation on the world stage.

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Agreed, horses for courses. In many areas Competition is a necessity, in some desirable and in some even counter productive. I believe in our case with the situation we have and considering the subject is of Key National infrastructure which is now actually becoming an essential utility as is wireless communication/broadband. To maximise the benefits of both and the heavy lifting high volume of Fibre enabling a viable wireless sector by load shedding. The best viable option for us is the path we have taken, fortunately it's brilliant fiscally prudent model minimises taxpayer cost over time rather than the massive subsidies model which maximises taxpayer cost , which in comparison is complete fiscal mismanagement on a grand scale
    Abel Adamski
  • We've now had many years in the Australian context to experience the coalition's 'private provider and light touch regulation' policies in practice, and they have not served us well.

    There is not enough market depth in this country to foster market-based competition beyond the urban centres, and in the absence of effective anti-trust laws the well known 300Kg gorilla is always going to do what gorillas always do.

    It will be a pity if Malcolm Turnbull attempts to brandish the experience of little Palo Alto as having any relevance to NBN in this country, because that will demonstrate that he has no understanding of the differences involved.

    It will take ten years to complete NBN, and by that time there is no question that the capacity and access available at that time will be fully utilised.
  • Palo Alto LGA (Local Government Area) (City of Palo Alto) consists of 61.858 square kilometres of land.

    To put that into perspective, Parramatta LGA (Parramatta City Council) is 61 square kilometres (with a river running through part of it).
    (Apologies if you're not familiar with Sydney)

    Anyone who tries to compare and then spout negativity about the roll-out of a network in Australia which is 7,617,930 square kilometres to an area that is 0.0008% the size is beyond idiotic!
  • yeah I don’t see how this rejection hurts the NBN in anyway, the way I see it if we do anything different to what Americans are doing then we must be doing something right at least when it comes to communications infrastructure, just think of all those FttP deployments all over the world and how due to America being the way it is it's virtually impossible for them to do the same thing. Americans will be looking at not only those deployments but also the stunning NBN in a few years with compete envy and be seething with rage... also consider that about 40% of all Americans still believe in creationism, do we really want communications infrastructure examples from such a backwards country?
    Hubert Cumberdale
    • I agree... because only a country with strong bashing of other peoples views could possibly be considered progressive.
      • AU > USA. Deal with it.
        Hubert Cumberdale
    • I agree... because only a country with strong bashing of other peoples views could possibly be considered progressive.
  • 100% agreed David. This, when you look at it, only serves to reinforce the whole debate around the NBN. If competition existed in Australia healthily via cable or even via copper, the NBN would be far from viable as a government infrastructure expense, let alone as a government business. And yet, it is viable for both.

    But as it stands, with the level of competition we have, the NBN makes the most sense and gives the greatest outcome to the consumer which is ultimately the point of the government stepping in. It would be fantastic if the majority of us had access to competition in the form of other provider's copper or cable in regional areas, but we don't. And while that stays the status quo, which seems likely with no change in the market, the NBN is the best way to break this lack of competition deadlock.

    Hence, as you say, the ACCC's allowing of the NBNCo.-Optus and Telstra deals. The ACCC recognises this deal as the best way for consumers to ultimately get competition back in their telecommunications services in the long term.
  • Taking Palo Alto as an example here doesn't work. This doesn't mention the fact that Palo Alto is one of the most affluent cities in the Silicon Valley, Steve Jobs was living there. Many rich geeks live there and most of the people working in High Tech want to live in Palo Alto. Palo Alto also has a lot of high tech companies (heard of the Xerox PARC? Xerox Palo Alto Research Center?). The project was started back in 2004 or 2005 I think, before the crisis. At the time, many of the affluent people were ready to pay $1000 for a very fast internet connection. Mention was not made either that there were other plans, comparable to ADSL and with better speed and/or cost. The conclusions and opinions seem a bit biased here. I lived in Palo Alto for quite a bit, I am not affluent and I would have subscribed to the fibre. To conclude, take a better example for a comparison please, or expose a bit more of the environment.
  • "Does Palo Alto's FttP rejection hurt the NBN?"
    No, because the NBN isn't rolling out to Palo Alto.
  • Imagine if they were in Australia, looking at at the NBN...

    The report considers a best-case scenario in which homeowners pay AU$0 to get a FttP connection, and then pay AU$24-36 monthly for their subscription. At a take-up rate of 100 per cent, the report concluded that such a system would gain several hundred thousand dollars per year.
    Paul Krueger
  • I simply cannot subscribe to the theory of infrastructure competition. How people can keep a straight face and claim that competition in infrastructure that is a natural monopoly is the most efficient way to build it is beyond me.
  • I'm for the NBN (ie for better broadband for Australian consumers) in principle, but I can't swallow your argument. You claim that HFC in Australia does not provide any worthwhile competition and use this as an argument for the NBN and FTTP. If HFC is not a competitive threat why is NBN paying Optus $800M and Telstra several billion not to compete? The fact is that near universal FTTP would not be commercially viable. To eliminate competition and ensure a high participation rate, Conroy and NBN have used legislative stick and financial carrots in ways that are blatantly anti-competitive, while the ACCC has had to look the other way. This is not the fault of the ACCC. Now there are social and community benefits expected to flow from the NBN, and the overall benefit, though difficult to quantify, may justify the means. But that was not your argument!

    Finally, David, here's a question for you: why does the NBN perpetuate the ADSL practice of providing such low upload speeds? Having fibre all the way in both directions there is no bandwidth restriction, and it should be possible to provide symmetric connections for modest incremental cost. With cloud computing upload capacity becomes really important. Are we missing a great opportunity here? Will we need NBN2 in a few years?
    • Achilles, you are correct that broadband must be all about upstream speeds, but how can you characterise the NBN's 40 Mbps upstream as a slow speed?

      The latest Senate Committee data (for April 2012) reveal that in REGIONAL Australia, 37% are actually choosing the top tier 100/40 service, and only 16% the 12/1 service. This means that 84% of all active NBN fibre services are running at upload speeds of 5, 20 or 40 Mbps, massively faster than the 1 Mbps typical of ADSL.
    • Achilles.

      NBNCo are paying Optus to migrate customers, for HFC pit and duct access (iirc) and as compensation for previous investments in actual infrastructure, which (even though arguably obsolete, because of fibre) will be made redundant through no fault of Optus’.

      This is what happens in business and those who keep bagging the NBN for being a GBE, can’t have it both ways. I bet the same people would be whinging if NBNCo just steam rolled over or seized Telstra and Optus' property 9and rightly so)?

      But shutting down old stuff and building newer, better stuff is commonly (commonly as in everywhere else on earth but the NBN) referred to as “progress”…!

      Tell me did you complain that you were forced from analogue to digital wireless? Analogue to digital TV? Dirt to bitumen roads? No, I bet you accept these as progress, but because the NBN is a political mine field, such silly comments are ridiculously being made and repeated.

      Gee I might start complaining about being forced to use copper, as I have no “infarstructure choice or competition”!

      And…people claim there isn’t infrastructure competition “because there ISN’T” and claims that the NBN will eliminate competition, is FUD!

      Seriously who cares who owns the (fixed) network? Do you care who owns the electricity cables? The water pipes?

      When you ring Telstra, Optus iinet, TPG or whoever, you want the best plan/data/speeds etc for the best price, it doesn’t matter who owns the network (different in mobile of course). The telling factor with NBNCo (fixed) is, that you are dealing with a wholesaler who is NOT a retailer (unlike Telstra) which is a far superior and fairer competitive environment for all RSP’s, creating better retail competition for consumers.

      It has been said (by those trying to bag the NBN) there will probably only be 4 or 5 truly Australia wide NBN retailers (RSPs) available to all Aussies. Err, as opposed to there currently being only one truly Australia wide retailer available to everybody!

      Then we have HFC from Telstra and Optus available only to (iirc 30% of Aussies) in a few cities, the rest (70%) are on slow, obsolete copper or worse. As opposed to 93% of Aussies, being on better than HFC and vast improvement in speeds and reliability for the remaining 7% via the NBN.

      So tell me which is better for consumers (not best for weird ideologues)?
  • Palo Alto is an insignificant red herring, better we look at the constraints to the HttP holus bolus folly.

    Fast fiber feed to Business, Health and Education is already in situ and is a must. Our aging population, where the young prefer mobility and the aged happy with HttN precludes the added borrowings to provide FttP.
    Vasso Massonic
    • The young prefer mobility, eh? Vasso, today practically EVERYOONE prefers mobility, but they still get 90% of their smartphone and tablet data off Wi-Fi that is delivered by fixed broadband. Mobile data is too expensive, because the physics of shared spectrum will always dictate that telcos must ration it by a high unit data cost.

      We have seen fifteen years of costly one-off installs of ISDN and DDE to cherrypicked premises like hospitals and government offices. The same needless cost is incurred if you don't supply FTTP to 20% of a suburb, just because it is connected to its twenty-year-old HFC service, which is already under strain in the evenings. This is why offering a customer migration subsidy to Telstra and Optus for their customers makes economic sense as part of the universal provision of the permanent fibre solution to all cities and large towns. It eliminates any delay to bringing the wholesale revenue from those customers into NBNCo to repay the construction cost sooner. Everyone wins, taxpayers, telcos and customers.
    • Err Vasso, it's no good one end, even in Health, Business and Education, having fast broadband when the other end they wish to be in contact with has but a tin can?

      Think about it!
  • Achilles - there is no need for a NBN2 in a few years - the 'natural' downlink:uplink ratio for the NBN GPON fibre technology is 2:1. The low uplink speed on the fibre plan of 12:1 is a product packaging issue, not a technology limitation. The next tiers higher of 25:5 and 100:40 are much closer to symmetry. Its only the lowest 12:1 plan that is close to the ADSL practice - and this is more because of a desire to have at least a baseline plan that was the same over fibre, wireless and satellite - and 12:1 is doable on wireless and satellite.
  • Palo Alto math 101: 26000 households x 20% connected x $5 extra per month = $312,000. So for 80 bucks instead of 75 they get fibre and cover the annual running cost. And of course if FTTP was offered, especially if it gave Americans a way to escape the bully oligopoly cable providers, expect the participation rate to skyrocket.

    Conclusion: There is no economic argument against FTTP to Palo Alto, just bulldust from incumbent telcos. Who funded the report?