Tuesday, October 31, 2006

EARNINGS SEASON://Second thoughts on KPN's Q3

Just as you think you cornered the VoIP market under a straightforward name (excuse me, we're leaving out a tiny segment, the peer-to-peer apps), your Big Competitor comes around and makes fun of you, morphing it into 'Internet + Bellen'.

Anyway, most striking I found the ease at which takeovers are regarded a form of SACs (listen to the call). Without Telfort, all these ISPs and ICT providers that KPN has been hoovering up, growth was squarely negative, instead of +3.7%. Also, there is no turning point yet in the development of the number of accesses.

As KPN managed to put an end to LLU and force competitors (they have until November 7) to rethink their strategy (turn reseller or invest in FTTC (as in 28k street cabinets) + SLU), munifiber is simply not talked about. Not a word about Citynet or any other project. When I spoke to the IR people recently, they quickly turned to Appingedam - as if that case is representative.

Thursday, October 26, 2006

CABLE OPEN ACCESS://Dutch decision raises many questions

The Dutch parliament's decision to impose open access on cable companies raises many questions. At first sight, it's a positive for KPN (it levels the playing field, KPN could be a service provider on MSO networks) and a negative for MSOs (new competition, new types of investments). The Netherlands are certainly not following the US model (no open access to either copper or coax) and it could be a precedent for European regulation.

It looks like cable open access is many years away. Still, leveling the playing field sounds fair.

Below, I have the translation of the bills that were passed (thanks to Citynet's Dirk van der Woude).

Questions raised include:
  • How will the EU respond? Is SMP relevant at all?
  • If line rental and services are to be separated, does open access concern (broadcast) TV only, or will it extend to (digital TV), BB and voice?
  • Will OPTA (or a different regulator?) go so far leveling the playing field demanding infrastructure-based competition, i.e. competitors installing gear at cable headends? Is this practical at all, from a technology point of view?
  • How will any costs of opening up the networks be divided among the MSO and the alternative service provider?
  • Will this allow UPC (Liberty Global) and Casema/Kabelcom/Multikabel (Warburg, Cinven) to merge, creating a near-nationwide network, further leveling the playing field?
  • How will Warburg and Cinven respond? Could the Kablecom takeover (from Essent) be in danger?
  • Longer term, is there room for three competing physical layers, now that munifiber is coming? Could it spur structural separation, all players using a single physical layer (owned by some state-owned body)?

Motion 1- Considering that the telecom sector more and more has strategies towards vertical integration of networks and services
- considering that this integration can hinder free competition between service providers
- considering that existing competion end telecom laws offer insufficient means to resist these competion barriers
- considering that government as well as parliament are in favour of a standing charge model (as proposed in the earlier accepted motion of MP Atsma of 2004)
- invites the government to propose within a year to parliament for changes in the Telecommunication law, by which the markets for infrastructure and services will strictly be divided by the introduction of a standing charge model as well as the prohibition of the conditional sale of network and services.
On the above motion the new Trade minister Wijn has said that it has all of his heart.

Motion 2
- Considering that developments in the telecom sector have been rapid,
- considering that there hardly is any difference anymore between telco and cable networks, form the perspectives of technology as well as use
- considering that free access for all service providers furthers innovation and ICT in the Netherlands
- invites the government to propose within a year to parliament for a change in the Telecommunication Law, leading to mandatory, open, non discriminatory access to all networks for all service providers.
Of this motion the new minister said that it might find problems in Brussels. He did not say that he disagreed with the gest of it.

B2B://Thomson preparing bid for Wolters?

Thomson Corp selling Thomson Learning seems to suggest Thomson is preparing a bid for Wolters Kluwer, if the latter's stock price is any indication. Wolters moved up by over 4% yesterday and closed up 2%, after the Thomson story leaked to the Wall Street Journal. Thomson wants to "take advantage of opportunities to accelerate the development of our core businesses and explore adjacent markets that are consistent with our strategy".

Also: Wolters put its own educational unit up for sale last month and Thomson reports Q3 results today. Riverdeep is rumoured to bid for Houghton Mifflin.

Thursday, October 19, 2006

VOIP://Skype data points from eBay's Q3 report

Data are limited in the report and the conference call. No active users or paying (SkypeOut and In, Voicemail) subs, the way eBay and PayPal are illuminated.
No word on Skype for Mobile.

What is there on Skype (consolidated from 05Q4):

  • 135.9m registered subs in >200 countries, + 23m = +20% qoq.
  • Revenues $50.0m, +13% qoq. Of this, 84% is from non-US v. 86% in the previous quarter.
  • The 2006 revenue target has been lowered to $195m (originally $200m).
  • Break-even is expected during 2007, "... while making significant investments to continue to build out the ecosystem and expand the product offering. (...) We're looking at a variety of incremental revenue streams ..."
  • Sky-to-Skype minutes: 6.6 bn or -8% qoq ("summer seasonal impact") and +77% yoy.
  • SkypeOut minutes: 1.1 bn, +32% qoq and +235% yoy. Partly on the back of free promos in the US, Canada and France.
  • The ecosystem comprises 3500 developers and 400 partners, including NetGear, Belkin and Sony (the cool new Mylo) for hotspot calling.
  • Skype for Pocket PC was downloaded 5m times.
  • Trials are ongoing of Skype Me buttons on eBay en Kijiji sites. "We see great promise. (...) we see indications of a higher conversion rate when there are Skype buttons that are used."
  • Skype fuels PayPal adoption.
  • Skype is part of the US deal with Yahoo! and the non-US deal with Google for click-to-call. Trials to start early 2007.
  • The 2007 guidance will be updated in January, depending on the results of several inititatives, including "Skype momentum to the fourth quarter."

UPDATE: lowering the revenue target will likely reduce the $1.5 bn earn-out (based on 2008 targets)

Tuesday, October 17, 2006

REGULATION://Copper access at KPN street cabinets

The issue of physical space at street cabinets has haunted the options for post-LLU competition for some time.

Last week KPN published SDF access tariffs, in repsonse to OPTA's positionpaper on KPN's All-IP network proposal. This space issue is still not resolved, though. OPTA's ruling is due late December.

KPN today tell me by mail:
"Het kan inderdaad zo zijn dat sommige kasten niet groot genoeg zijn, echter als de unbundlers vroegtijdig aangeven om hierin ook te willen investeren kan dit gelijk met KPN worden meegenomen."
Which reads like:
"It can indeed be the case that some street cabinets are not large enough, however as the unbundlers indicate early to also want to invest, this can be incorporated by KPN."

Still not resolved, then. How many street cabinets does this concern? Will KPN have to pay up, or would the costs be shared? etc.

Monday, October 16, 2006

DIVERGENCE://Business v. consumer services

Among all the convergence going on (IP/IMS, FMC, UMA, dual-mode handsets, telecoms/IT, telecoms/media, software-defined radio, eBay/Skype, etc.) there is still a sense of divergence between corporate and consumer services.
  • Consumers: DIY, user-generated content, web applications.
  • Businesses: focus on core-business, (outsource) ICT-services.

Hence telecoms operators like KPN and BT trying to focus on a low cost high-bandwidth network for consumers and at the same time trying to deliver ICT services to corporations.

No wonder KPN would like to buy an asset like Getronics. From a quality point of view, other candidates may make more sense, e.g. LogicaCMG (which seems to be struggling building international coverage) or Atos Origin.

Thursday, October 12, 2006

TRUCK ROLL://Cable guy in action

The truck roll is a large part of opex, but I suspect a different kind of roll here.

Wednesday, October 11, 2006

ACCESS://First, second, third, fourth ... pipes

There is a plethora of choice in access land (to support the low cost operator stance from Mr. Enck). Check out today's news below.

It makes investment decisions very hard.

  • Fixed: Is FTTH really the best, or can FTTN + DSL beat it? What about HFC, BPL and BiG? Is it wise to make big bets at all right now (FiOS), or is it better to be be slow?
  • Wireless: Will any wireless technology ever be able to compete? How about xMax? Is it wise to make big bets now, or try to be slow and try all technologies (Sprint)?

Today's news:

  • DSL: ECI Telecom will lead a consortium aiming for higher speeds, using dynamic spectrum management (DSM). Beyond VDSL2, mind you! John Cioffi argues copper has more available bandwidth than fiber.
  • Cable: a comparable view, saying cable plant has plenty of bandwidth. However, with DOCSIS 3.0 still in development and DSL racing beyond VDSL2, this seems a tough point.
  • BPL: Fiercewifi argues perhaps the time has come for a breakthrough. We have to look out for the Microsoft EMEA event in Munich in November to see a Motorola demo.
  • TDtv: The British wireless incumbents will do a trial of IPWireless' TDtv solution, using existing spectrum. Jointly, that is, a smart decision. However, DVB-H seems to be the winner so far in Europe, and there is plenty of choice; like:
  • MediaFLO: Sprint Nextel has a trial going on. Why not? They have CDMA-EV DO-Rev A, iDEN and (to be launched) WiMAX, so why not add a fourth network (not counting the possibility of yet another from the SpectrumCo/cable joint-venture. This would be something like: we try to optimize the utilization rate of our towers.
  • Picocells: T-Mobile will use ip.access' nanoGSM gear for in-building coverage.
  • Ruckus Wireless: Belgacom will use this company's WiFi solution for in-home networking, to support the Belgacom TV (IPTV) product. Ruckus just received funding from T-Online Ventures (a T-Mobile sister in the DT kamp) and Motorola.

Friday, October 06, 2006

DISRUPTIVE INNOVATION://BT Contact, xG IPO, Venice Project, Jajah Mobile

I've done it again: put some possibly disruptive innovation together in a post on my Dutch language blog. How's that for disruption!
Just follow the links (or use a translation tool, see right).

Wednesday, October 04, 2006

MOBILE SUB-BRANDS://E-Plus' strange vybemobile offering

KPN-subsidiary E-Plus went one step further down the sub-branding road, which is working out very well for them.

I have a hard time understanding the new vybemobile offering, though, for three reasons:

  • It focuses on the youth market, but so do Simyo and Base, to an extent. Specifically, it targets music lovers, but so does Viva Mobile (with MTV).
  • Vybemobile is a venture with Universal Music. That makes it appear limited. What about Warner Music, EMI and Sony BMG content? Or will E-Plus add more brand? That would leave us no alternative but to buy 4 new handsets! Again, isn't Viva enough? Or maybe the service should be a licensing product, the way the new Mobile ESPN will work.
  • KPN wants to get out of subsidizing handsets, but this service seems to need it.

HDTV://Predictions for 2007 - kick-off

This must be among the first of the usual torrent of new year predictions. Philip Swann has a pretty good hit rate.

THIRD PIPE://Citynet to start on FTTH in Amsterdam

Citynet will officially start work on the Amsterdam FTTH-network next week, October 12. The first phase targets 40k homes.

The EC has not yet formally approved the venture, but Citynet regards a court ruling supportive. Moreover, they made sure to follow the European Market Economy Investors Principle. Citynet also refers to other state-owned firms providing BB access (e.g. MSO Zeelandnet).

  • Owner:GNA (Glasvezelnet Amsterdam), which is in turn 33% each owned by the city, 5 housing corporations and several investors (ING and Reggefiber, which we also know as Lijbrandt Telecom owners).
  • The network will be operated by bbned (= Telecom Italia), which rents the network from GNA, on a wholesale basis.
  • There will be many service providers.

REGULATION://Post-LLU competition

Returning once more to OPTA's proposal for post-LLU regulation. It is interesting to see how different comments are. Having no English version available, does not exactly help.

Some comments concentrate on delays in closing the old network, others report KPN's initial positive stance.

My takes:

  • Positive for KPN is OPTA's acknowledgement of the rationale of the new All-IP network. It may be built and KPN will be allowed to close down 1100 COs.
  • Positive for altnets is OPTA's commitment to infrastructure-based competition on the telco network. But who would have expected an American style duopoly at this point?
  • Something that seems to have gone unnoticed is OPTA's plan to study the British model (equivalence, Openreach). This study will be conducted shortly and will be incorporated in the final ruling (late December).
  • OPTA seems to urge altnets (and others?) to cooperate. Enter speculation on Australian style initiatives.
  • The final ruling could have major impact across Europe.

UPDATE: KPN made an English translation of the OPTA paper. Mail me for a copy.

Tuesday, October 03, 2006

REGULATION://What comes after LLU in the Netherlands

OPTA today issued a Positionpaper All-IP, responding to KPN's plan to build a FTTN/VDSL2 network. It is all about phasing out LLU and coming up with a fuklly fledged alternative. Interested parties have 4 weeks to respond. OPTA aims to have new regulation place by late December of this year.

First, the facts from the Management Summary and the paper itself:
  • KPN must be allowed to build the network and close COs as planned, but negative effects on competitors must averted.
  • Infrastructure-based competition is OPTA's primary goal. Only wheneven it is impossible will OPTA resort to services-based competition.
  • OPTA calls upon altnets to define alternative strategies.
  • KPN's planned exit of COs (and thus MDF access and the sale thereof) must be accompanied by a migration strategy for LLU operators. LLU is to be replaced by a new form of infrastructure-based competition. OPTA aims at altnets rolling out their own FTTN network (from COs to street cabinets, i.e. KPN's Metro Access Network, which in fact is for large part completed), facilitated by KPN. OPTA sees a range of possibilities: from renting capacity on this MAN (SDF backhaul) to sharing the trenches. However, OPTA also recognizes that probably only KPN has sufficient scale for the latter investment.
  • Closure of COs will be restricted in order to allow LLU operators to recoup their investments. OPTA sets this period at 5 years. In addition, the timeframe would be such that KPN needs to notify the market of closure of each CO 2-2.5 years ahead, after which LLU operators have 1-1.5 years to migrate.
  • KPN is held to provide service-based competition for areas that are not yet covered by DSL and that it plans to add.
  • OPTA will examine the British model (equivalence, Openreach), especially should KPN and altnets be unable to work out a model on their own.

OPTA will formulate a full alternative to current MDF access. It will likely include a choice from these options:

  • Access to street cabinets.
  • Conditions for the phasing out of MDF access.
  • A regulated wholesale broadband product form KPN wherever it offers no SDF backhaul.
  • Dark fiber access or regulated SDF backhaul, plus co-location at both street cabinets and MCLs (higher up in the network than current MDF locations).

KPN's All-IP plans include reducing the number of COs from 1350 to 200, roll-out fiber to 28k street cabintes and deploy DSL (including VDSL2) in the last mile. Build-out is to take from 2006 through 2010. Selling-off real estate should furnish EUR 1 bn or at least half the cost. Some 8k jobs (of 20k) will be lost.Opex savings are targeted at 850m EUR/yr from 2009.


  • OPTA opens a consultation phase, lasting 4 weeks. It could potentially be a long process, but roll-out by KPN should not be frustrated too much. At the same time, KPN is very smart by having a large part of the work already done. By the way, this means that responses are due October 31, the day KPN reports Q3 results. OPTA aims to have new regulation ready by late December 2006.
  • OPTA seems to call for action from altnets. The regulator is quite explicit about the end of LLU. Maybe they hope for a Free/Iliad or A9 (Australia) style initiative. So far, we have seen little action, most notably all of them allowing KPN to acquire a string of ISPs (OK, LLU investments are at risk, but should be recoverable). Why is it so hard for those (cash-strapped) competitors to work together, possibly even with municipalities?
  • A limited negative for KPN is a delay in the planned closure and sale of COs.
  • OPTA's commitment to altnets and infrastructure-based competition is positive for competitors, but should not surprise a soul.
  • Possibilities for co-location at street cabinets remains unclear. OPTA seems unimpressed about any problems here (SDF-co-location). KPN places its NG-DSLAMs at these street cabinets to form MSANs (multi-service access nodes).

Monday, October 02, 2006

REGULATION://Tomorrow is D-Day for KPN and altnets

Tomorrow will see OPTA's assessment of the future of copper-based competition in the Netherlands. A very significant event for KPN, altnets (Tele2-Versatel, Orange/Wanadoo, bbned/Telecom Italia) and cablecos (UPC, Casema/Kabelcom/Multikabel).

  • KPN is building its All-IP Network (FTTN + VDSL2) and the tiny problem of limited space in street cabinets (too small for colocating LLU-players).
  • KPN acquiring a host of ISPs, lastly Tiscali NL.
  • Warburg Pincus and Cinven trying to merge Casema, Kabelcom and Multikabel.
  • KPN's failed attempt of building a national FTTH-network together with cablecos and the latter's belief in the future of HFC-networks.

Which are the possible outcomes and OPTA's range of choice?

  • Adopt KPN's choice: just resellers, no LLU, on the All-IP network.
  • Force KPN to add a street cabinet at each location: a second one for colocation. I'm not sure this could work.
  • Structural separation, to any degree (ranging from Telecom Italia's new strategy (?) or BT's Openreach to Portugal's condition for the PT acquisition by Sonaecom).

Judging from what happened in Germany (no regulatory holiday for DT's FTTN-network), an American-style market (duopoly) seems unlikely. Besides, OPTA is too much in the habit of slamming KPN. If structural separation is proposed, it will shake European markets. So, OPTA must be studying on some kind of 'polder model'.

Finally, there is still a chance for a superior national FTTH-network, either from KPN or from any combination of players (e.g. Australian-style), but that will be hard for OPTA to have a word in.