KPN's biggest fear: cable starts competing on speed

Monday 11 May 2015 | 14:05 CET | Background

The earlier announced slowdown in KPN's roll-out of FTTH was underlined again in the first quarter. The operator added just 45,000 homes passed, the smallest number since 2011, when Reggefiber was still working on its roll-out model. KPN's focus has shifted, at least for the moment, to the copper network. This is a risky strategy. Cable operators may disrupt KPN's plans by turning to faster broadband speeds and investing in Docsis 3.1 technology. Given Liberty Global's leverage, it can't afford to stand still; Ziggo must grow. A few years ago KPN's ADSL services lost ground to cable after the roll-out of Docsis 3.0. There is a clear risk that history may repeat itself, as KPN's VDSL strategy fails to hold out once the cable sector starts on Doscis 3.1 and phasing out analogue TV. 

KPN is working to make faster broadband available nationwide, and once it completes its copper upgrade, the focus may return to FTTH. A number of techniques are possible over copper (VDSL2 from MDF or SDF sites, pair bonding and phantom mode, vectoring). Some of these are still under development, such as G.fast and innovations from Alcatel-Lucent (Vplus), Huawei (SuperVector) and Adtran (FDV). KPN aims to offer at least 40 Mbps to 90 percent of households by the end of 2016 using VDSL2. Speeds of up to 100 Mbps (VDSL2 + vectoring) should be available for 85 percent, 200 Mbps (VDSL2 + vectoring + bonding) for 70 Mbps and 500 Mbps (FTTH) for 33 percent.

KPN and its unbundling customers have been rolling out the new technology since 2010. Tele2 was the first to offer VDSL from existing MDF sites. Despite all the upgrades though, the DSL market continues to contract, in favour of cable and especially FTTH.


The question is whether KPN has adopted the correct strategy. Are the expected speeds over copper realistic? Wouldn't it be better to pursue both technologies and continue the FTTH roll-out unabated? And what will cable's response be?

Investors are clearly not worried given the recent increase in KPN's share price and the operator's decision to shrink its investment budget. This assumes the network will be able to stand up to the competition and more room is created for paying dividends.

Nevertheless, there are risks associated with KPN's strategy. First off, the company is using experimental techniques. Copper technology depends to a great extent on the customer's distance from the exchange, and the actual speeds may vary considerably from theoretical aims. Alcatel-Lucent said itself recently that "VDSL2 and vectoring alone can't achieve 100 Mbps reliably over copper". Another problem is the asymmetry: the upload speed is much less than the download. FTTH doesn't have this disadvantage. 

There's also the costs. These are not just a fraction of FTTH; we're still talking about hundreds of euros per connection. KPN said at its Q1 2015 results that upgrading to FTTH later is simple and requires "merely" some excavation of a few hundred metres. This still results in considerable costs. It's also very likely that if KPN does upgrade to FTTH this way, it will use GPON technology, due to its lower cost.

KPN is promising investors that investment costs will decline. It's highly questionable whether KPN will be ready and willing to raise its spending again after 2016. 

Liberty Global is still working on the integration of Ziggo and UPC. The market may fear the emergence of a duopoly in order to "optimise" (ie, lower) capex. Liberty Global has a much more leveraged balance sheet though (it targets a debt ratio of maximum 5.0 times EBITDA versus 2.5 at KPN) and is dependent on continued growth to sustain its financing. More likely Liberty Global now smells blood, after KPN chose copper and the quick introduction of faster speeds. It could roll out Docsis 3.1 technology and shut down analogue TV quickly in order to free up more network capacity. Cable infrastructure is less robust than FTTH, given the shared bandwidth, but can still offer more than DSL networks. Demand for higher speeds is still limited, but is growing. Offering 'superfast' broadband or 'the fastest broadband' also has a definite marketing value. Data traffic is growing quickly, in part from the mobile sector (80% of mobile traffic goes over Wi-Fi to fixed networks). 4K TV may soon be a mass-market reality as well.


Few subjects in the telecom sector generate as diverging opinions as the 'need for speed'. In the end it's all about timing - everyone agrees that in the long run all networks will be completely fibre. The only question is when to embrace the upgrade. Operators investing too soon end up with an empty network and low return on investment. Those who wait too long lose market share. The issues to consider:

  • Demand for 'superfast' broadband is still low, but is growing. The rise of 4K TV (ultra HD, UHD), the Internet of Things (IoT) and Wi-Fi offloading are uncertain factors.
  • Operators have an incentive to switch to FTTH: maintenance costs (opex) are considerably less than for copper.
  • KPN is a listed company with shareholders that have a shorter-term vision than what may be required for this type of network project. This means the lower costs of FTTC may be welcomed sooner than the high investments required for FTTH.
  • Rolling out fibre is a costly and long-term project. Investing in copper is unavoidable, even if it is considered an interim strategy.
  • The process of rolling out VDSL (FTTC) now and FTTH later can be optimised by making the FTTC network 'FTTH ready'. KPN said the costs per household for FTTC are around EUR 250, versus EUR 800 for FTTH. The cost of upgrading FTTC to FTTH should be somewhere in between those two amounts. If a suitable network topology is chosen, the operator can avoid a large part of the FTTC costs becoming 'regret investments'.
  • In terms of performance, KPN doesn't have much to fear from cable when it comes to its FTTH network. It's the DSL base that cable is attacking. It appears likely that cable infrastructure will have more capacity in the coming years than the copper network, but this isn't for certain. Cable is shared infrastructure, and Docsis 3.1 hasn't reached the market yet.
  • Last but not least: other factors than speed play a role, such as price, TV quality, TV innovations, content etc.

History to repeat itself?

Investments in broadband are rife with uncertainty, with much dependent on timing. KPN faces a risky situation if Ziggo, once its integration with UPC is completed, goes in search of growth. KPN may lose further market share to cable, after already suffering in recent years from the introduction of Docsis 3.0. The upgrade to Docsis 3.1 can give the cable operators a new weapon to take more customers from the DSL market.

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