Will Telstra bid if it is structurally separated? Betcha!

Our fourth poll has ended, but to no obvious conclusion. A ridiculously small sample was heavily skewed toward structural separation (65%), but functional separation wasn't completely off the table (35%). Operational separation (5%) and accounting seperation (which you couldn't even vote for!) are excluded going forward.

The topic gets a lot of press these days, mostly in New Zealand, Australia and Italy. Check out this Arcep document for an introduction to separation.

Here is why we believe in structural separation.

• In a world of intramodal competition on the telco network (and intermodal competition between copper and coax) there will never be full equivalence between all players (incumbent, unbundlers, resellers, etc.). As much as PTT's want symmetry between telco and cableco competition (i.e. open access to cable networks), they should also allow for symmetry on the copper network. There will never be true symmetry if one service provider also owns the network, and the others don't. No matter what they say, the incumbent will always be at an advantage.
• Look at it from a synergy point of view. The incumbent reaps all the synergy benefits stemming from owning both the network and a service provider. These advantages should be equal and shared. And hence, all incumbents cry foul when confronted with the threat of being structurally separated. But that's the whole point, brothers and sisters: the regulator should focus on simply making PTT's smaller and creating long-term competition from viable altnets.
  
• KPN has successfully staved off structural separation. On the one hand, this is due to its full portfolio of wholesale services and a certain co-opetitive stance toward resellers. On the other hand, the world is facing next-generation access investments (i.e. FTTH), which are not only expensive (to be carried by a company the size of the incumbent only) but also create huge regulatory uncertainty. In the Netherlands, hardly anybody is left to consider serious and long-term competition (apart from cable). Orange Broadband is now owned by T-Mobile and put up for sale; bbned and its sisters can hardly be taken seriously because parent company Telecom Italia has a lot on its mind; and Tele2 seems to be withdrawing from western Europe altogether.
• Only when structurally separated can the telco appeal to the right investment communities: the dividend aficionados can buy the network, retail minded investors can focus on service providers (higher risk/return profile), etc. Also, only in this way can the network attract subsidies or create public/private partnerships. In a way Telstra acknowledges this: if it is structurally separated, it will not bid for the National Broadband Network contract and subsidy (AUD 4.7bn). But once separated, the network company will surely bid for the contract; I will eat my hat if it doens't!
  

We have looked very hard and closely at all the arguments against structural separation, but none really seems to make sense. Yes, it will be quite disruptive. And it will costs a few pennies. But to say that it would take away any incentive to invest just isn't very 21st century thinking. Finally, to say that you want to own the network (to generate the cash and allow you to pay fat dividends) is not very clever in light of the above (we want equivalence and symmetry, right?).

To round off, we want to share some fun related to the topic - unless it brings you to tears, of course.
Telecom Italia is one of those companies that may face structural separation and it will come as no surprise that Tiscali is all in favor. FastWeb takes a different position: they think it's a bad idea! Functional separation would suffice. But wait a second: isn't FastWeb 82% owned by Swisscom, another PTT?
With hindsight, that calls for a round of applause for Optus (the Australian subsidiary of SingTel). It openly called for structural separation of Telstra, even if it's parent company was fighting the same fate in Singapore. Or is it the other way around: was SingTel being a hypocrit?

Silicon Valley business model in the Netherlands: be bought by KPN

The KPN takeover was looming large over the Tiscali Wholesale Annual Partner (ISP customer) Day I was invited to speak at, last Thursday. OPTA (the Dutch NRA) has spoken supportively of NMa’s (the Dutch competition watchdog) approval of KPN buying Tiscali NL. Closing is expected this month.

I suppose KPN’s plans could be anywhere between fully dismantling Tiscali NL and maintaining it as a standalone operation. One thing seems inevitable: a namechange, as Tiscali SpA’s Italian and UK operations are now completely unrelated. Possibly KPN will have the retail branch focus on a certain demographic, under a new label. The wholesale operations could actually be kept at an arm’s length to service ISP’s who suffer a certain degree of ‘green-phobia’ (green being the KPN corporate color – formerly that is). The latter part of Tiscali NL will be cherished, I suppose, because it only recently scored its largest customer win: Vodafone NL, which is working toward it's fixed/mobile strategy. I hear that Vodafone issued its RfP around the time that KPN and Tiscali came to an agreement (September 2006), so when Tiscali NL was chosen, Vodafone was very well aware of the KPN takeover.

Vodafone NL was also invited to speak. Daniel Nordström presented the new fixed/mobile strategy. I wonder if Vodafone’s mobile portfolio could be offered to Tiscali Wholesale’s other ISP customers. Of course, that would be a tough sell at the new parent company (KPN), but it could be a smart way of luring green-phobic customers and at the same time get some inside information on a competitor. However, I doubt KPN would allow Tiscali this level of independence.

What stood out during the day was Tiscali Wholesale’s focus on the end user, which coincided nicely with my own presentation, which related Timothy O’Reilly’s Web 2.0 to STL’s Telco 2.0. People at Tiscali are definitely aware of those developments in the marketplace.

Also, I touched on LLU in the Netherlands, which is coming to an end because of KPN’s ‘All-IP’ plans. What will OPTA come up with to replace it (a ruling is due this month)? If SLU is not feasible (as Analysys calculated), something rather big seems to be heading our way. I believe LLU and SLU are intermediary strategies, fiber is a natural monopoly (both in the backbone and on the in-home level) and thus open access FTTH is inevitable. How about some separation (to add some recent news: Sweden is looking at the Openreach model)?

I aired my growing surprise at the lack of willingness to cooperate among network operators. Recent developments are few and include Australia (G9), possibly Sweden (Telenor/Tele2/Telia) and Nigeria (25 ISPs teaming to build a WiFi network). Sure, these operators are backed by competitors, but if you want to compete with the powerful incumbent, you better get together.

Which brings me to my final observation. We all know the ‘business model’ of many a start-up in Silicon Valley: be bought by Google. Now, somebody confided to me that a similar model was explicitly chosen by many ISP’s in the Netherlands: be bought by KPN. And we all know how that ended!

Come and meet me in Maarssen (NL)

Those wanting to meet the writer of this humble blog, all you have to do is become a Tiscali Wholesale customer in the Netherlands (now a KPN subsidiary). Next Thursday, June 14, I was asked to provide a sector view at Tiscali's annual 'partner' meeting.
More about this after the meeting.

Network sharing versus intramodal competition

As always, the TMT markets are moving rapidly. The interesting thing is that several seemingly unrelated deals can quite easily be connected. Let me set out on this small journey by starting with an Australian newspaper group and ending in the very same country.


The WSJ proves its value

The Wall Street Journal once more proves its value, this time at a very convenient moment. The Bancrofts may be pushing Murdoch for raising his Dow Jones offer, or they might want to entice somebody else to mount a counter offer.

Just two days ago a WSJ story carried this headline: Will Vodafone Be Put in PlayBy ABN-Energized Activists?. And today, John Mayo steps forward with his ECS Assets vehicle to push for the freeing up of up to GBP 38bn.

Bravo WSJ.


Will wireless be a duopoly market as well?

Further, one may question the long-term chances of a standalone wireless operator such as Vodafone, along the lines of consolidation in the European broadband markets: AOL has vanished (as an ISP), Tiscali has retreated to the UK and Italy, Pipex is up for sale, and today France Telecom and Deutsche Telekom are swapping assets (see below).
Vodafone's break-up value could be considerably more than its current market value. For now, mobile is a far more attractive game than broadband, in terms of margins, justifying 3, 4 and 5 player markets (not to mention markets like Bangladesh, served by 6 operators). But in a few years, mobile could be another utility. Look out for cablecos to snap up mobile operators once they have their networks and balance sheets under control.
Also, imagine the kind of cost savings when access networks are shared. In the end, base station networks are extremely overlapping access networks, which may very well be shared.


Telco/cableco duopoly nearing in the Netherlands

As to the FT/DT swap: today Ya.com is snapped up by France Telecom and Orange NL goes to T-Mobile.
Spain is consolidating:

• three major players (Telefonica, France Telecom, Vodafone)
• two standalone operators (Yoigo in mobile, Jazztel in fixed)
• cable company Ono.

The Netherlands too:

• it will be a 3-player mobile market (KPN, Vodafone and T-Mobile).
• there is no obvious buyer for the Wanadoo-part (ISP) of Orange NL.

What could happen to the former Wanadoo-part of Orange NL?

• I am sure T-Mobile is not interested - unlike FT, DT has a mobile-only strategy 'abroad'.
• KPN is restricted, as the regulator barely allowed its recent Tiscali NL takeover.
• Tele2/Versatel could be a serious candidate; if not, do not be surprised to see Tele2 abandoning the country altogether.
• Scarlet could step in and move to a facilities-based business plan (following the Tele2 example).
• Vodafone doesn't seem interested, as it uses Tiscali NL as its broadband partner for its 'Total Communications' strategy.
• Finally, bbned (Telecom Italia) is a candidate, but its commitment to the Netherlands is doubtful.

In other words, the Netherlands seems to be advancing toward this telco/cableco duopoly.

I believe PTT's are increasingly proving to be winners, but so far this seems to be visible in the Netherlands only. Pushing fiber deeper into their networks (FTTC: fiber to the street cabinet - not to mention FTTH) makes LLU a thing of the past and effectively forces altnets to follow or die (see below).

Unless PTT's allow munifiber to get a too strong foothold, will many markets move toward a US-style duopoly of telco v. cable.


If SLU is impossible, network sharing should be considered

As I have written several times before, KPN is trying to kill LLU by moving to an All-IP network, which includes FTTC. The most obvious replacement would be SLU (sub-loop unbundling: from the street cabinet, instead of the central office), but Analysys has shown that this is not economically viable.
OPTA, the local NRA, is grapling with this dilemma. A 'full alternative' to LLU was promised for Q2, but so far hasn't emerged from OPTA's offices.

In my view, the obvious way out would be the Australian way, where 9 altnets have come together to propose an alternative to Telstra's fiber plans. Sure, altnets are backed by competing companies, but shouldn't they set aside their differences to work locally on a country-by-country basis? It simply makes no sense for 9 altnets to want to each compete with a strong incumbent.

Still, one may question the long-term viability of intramodal competition (operating active elements like DSLAMs on the incumbent telco network). Fiber will be extended - sooner or later all the way to the user: FTTH. Any xLU model (LLU, SLU, ?LU) would imply altnets replicating more and more of the incumbent's network, in the end actually replicating the whole thing, as the copper last mile gets shorter and shorter. The only way out seems to be to sooner or later admit that a single fiber network is the only economically viable situation. Now that LLU is coming under strain seems to be the time to acknowledge this. In other words, altnets should aim for network sharing with incumbents. However, the PTT could be tempted into wanting to go it alone. Hence, an NRA like Arcep (France) is trying to facilitate network sharing.

The viability of intramodal competition in general was questioned earlier this week by the Australian Kevin Morgan. He referred to this an arbitrage game. Frankly, I hadn't looked at it that way before, but I suppose he has a point as altnets are merely kept alive by regulatory intervention. It reminds me of what the Bells over in the US kept repeating a few years ago, when the 1996 Telecom Act was replaced and intramodal competition was effectively killed: after 8 years of competition and cherry-picking, altnets should have built their own networks. However, Kevin does not acknowledge one important thing: how LLU operators have increased competition, driving prices down and broadband penetration up.

Anyway, the end of intramodal competition seems to be nearing. FTTH, fiber in the last mile, seems to be a natural monopoly. If PTTs do not see this, newcomers (like Reggefiber, Iliad, neuf Cegetel) will.

Which leaves cablecos: will they follow?

CONSOLIDATION://Orange NL for sale

A newspaper in the Netherlands reports France Telecom has hired Lazard to shop Orange NL around. The unit has 600k BB subs and 2.0m mobile subs and should bring in EUR 800-850m.

Observations:

• Fixed NL: KPN is forcing a choice upon altnets, building its All-IP network: retreat (or be a reseller) or step up investments to make SLU work, as the old paradigm (FTTEx + ADSL2+ = LLU from up to 1300 exchanges) is replaced by a new one (FTTN + VDSL2 = SLU from up to 28k street cabinets). The outcome is still up in the air, as OPTA seems to be backtracking on earlier support of KPN's plans, but Orange seems to think neither option is very attractive. Telecom Italia, through bbned, and Tele2/Versatel seem to be committed to the Dutch market.
• Mobile NL: The mobile market is going to a three-player model if T-Mobile or Vodafone moves in. T-Mobile could even go from a mobile-only strategy (as in the UK) to a triple play offering (as in France). Unless of course China Mobile, Weather, Telefonica, Telecom Italia, Belgacom, Swisscom, CPW or TeliaSonera (cf. Xfera in Spain) deems the time right for a new market entry.
• France Telecom: Going from 5 to 4 countries for its triple play offering (France, UK, Spain, Poland).

Consolidation is continuing, driven by a need for scale economies in mobile and LLU (not to mention SLU). More units could be put up for sale (DT France, DT Spain, Tiscali UK, Tiscali Italy, SFR). What is intriguing is:

• Companies are abandoning saturated markets, like Scandinavia and now the Netherlands (the Tiscali NL sale to KPN is pending at the NMa) and are turning to emerging markets.
• Vodafone, according to the newspaper, wouldn't be interested in Orange NL. Puzzling. Could this be the first step of Vodafone putting even more focus on emerging markets?
• If T-Mobile isn't interested, they might as well leave the market altogether.
• Telefonica is definitely a consolidator. Will they bid for Orange NL, or target the bigger prize: KPN?
• Will this be the European entry of an 'eastern' company (after Hutch and Weather)?
• As PTTs are fighting each other in their home markets, could this sale mark the formation of a pan-European kartel, e.g. FT and DT getting out of each other's markets?

UPDATE: obviously, existing players could strengthen their current presence:

• Telecom Italia: to add a retail business (and mobile) to their bbned offering.
• Tele2: to gain scale and a mobile license, in order to migrate their current reseller business.
• Scarlet: to mirror Tele2's strategy of turning into a facilities-based operator.
• Cableco: to add mobile and to sell the fixed business to somebody else.