Should KPN push ahead with FTTH or switch to VDSL?

Friday 10 September 2010 | 13:13 CET | Market Commentary

KPN announced at its half-year results that it would stop the further roll-out of VDSL to SDF locations (street cabinets). This suggests it’s abandoning FTTC. The Dutch operator will continue with the roll-out of VDSL to MDF sites (central offices), also known as VDSL@CO. In addition to ADSL, which will continue to operate for some years, KPN is developing FTTH, which is being rolled out by its partner Reggefiber. KPN’s choice for VDSL@CO is completely logical, as we said earlier when Tele2 pioneered this concept (see our commentary 'Tele2's VDSL roll-out a no-brainer'). It’s a relatively cheap way to further exploit existing assets (MDF sites with fibre backhaul). In contrast, FTTC is costly due to the need to take fibre to 20,000 street cabinets. It’s also no interim strategy to FTTH, as the Reggefiber network architecture is completely different and can’t elaborate on FTTC.

That’s KPN’s story. Recently market researcher Analysys Mason issued a notable report, giving its preference to VDSL over FTTH. While we have not seen the entire report, a summary of it says the conclusion is based on the success of mobile broadband, alongside the “troublingly low” take-up of next-generation access services. How does this idea square with KPN’s plans?

To start, we can conclude that mobile broadband has attracted more traffic and attention than was originally expected. The iPhone and other smartphones, apps, the iPad, Android – these are all driving growth in mobile broadband. While it’s no complete substitute for next-generation access networks (whether they are VDSL, HFC or FTTH), it’s more than just complementary. This doesn’t take away from the fact that there are still some services that can hardly be used on a smartphone, such as over-the-top (OTT) video. In addition, mobile broadband is a stimulus to roll out NGA networks, namely to give the user access via Wi-Fi. In other words, the growth in ‘connected devices’ is a driver for NGA. Furthermore, mobile and NGA networks can share their backhaul.

Also noted are the constant complaints of the lack of services making a NGA a necessity. This is a chicken-and-egg issue: such services will not be possible until the NAG is available, and a NGA will not get the financing until such services are available. Only a long-term vision can break this impasse, and it’s increasingly questionable whether operators have this kind of vision (see our commentary  ‘Cherry picking by operators is rational, but will encourage govt intervention'). There is clearly a role for government here, either to finance the NGA or to stimulate the development of new services. And there is actually already at least one service to justify NGAs: OTT video.

From the end-user perspective, there are still few signs that a NGA is needed. Last year we published a report that showed only 900,000 households (of the 7.3 million in the Netherlands) expected to upgrade to at least 50Mbps in the next four years. (see ‘Dutch Consumer Connected’, September 2009). Furthermore, surveys in the Telecompaper Consumer Panel show that consumers would sooner choose for the same broadband speed at a lower price than for a higher speed at the same tariff. In short, the increase in broadband speeds seems to have reached a plateau. In the Netherlands, Zeewolde is currently the champion with 200 Mbps (over FTTH), and there is not a single indication that the number two, the cable networks with 120Mbps, feel the need to surpass Zeewolde.

Without seeing the contents of the mentioned report, the authors appear to be standing in the shoes of the operators, which again are not planning for the long term. Price pressure leads to margin erosion and that makes NGA investments difficult to justify at the moment. Only a long-term vision recognizes the trends that justify a NGA network, with the long time needed for roll-out an argument for starting deployment already today.

Going back to KPN, we can see another factor at play – something that’s not relevant in every country, but is in the Netherlands: a cable network with nationwide coverage. As noted above, VDSL@CO is the more logical choice over FTTC, but even with that, less than half of the population of the Netherlands will be reached. Furthermore, the speeds are less than those of cable broadband. Our new Dutch Broadband Q2 2010 report shows that the DSL providers are increasinly losing subscribers to cable and FTTH. In other words, the presence of cable makes the roll-out of FTTH a sound, logical choice for KPN.

Free Headlines in your E-mail

Every day we send out a free e-mail with the most important headlines of the last 24 hours.

Subscribe now

::: add a comment