Cable Tech

Telcos Consider the Splits

It looks as if BT Group plc (NYSE: BT; London: BTA) and U.K. regulator Ofcom have started a trend, as carriers and regulators in Europe and on the other side of the world look set to follow Britain's example of structural separation for incumbent fixed-access networks.

BT's access network division is called Openreach (or Openretch, as it has become known in U.K. telecom circles, the saucy japesters), a distinct business unit with its own management team that was formed in September 2005 as part of an agreement with regulator Ofcom. (See BT Opens Up Access.)

Simply put, this creates a divide between the carrier's access network operations and its retail services division, which (in theory) has to deal with and buy services from Openreach under the same terms and conditions as do the U.K.'s alternative operators.

Now other countries are moving towards that same scenario. Just this week Sweden's National Post and Telecom Agency (PTS) staked its position in a "Broadband Strategy for Sweden" document that puts pressure on Telia Company to create a separate access unit: "PTS is of the view that the most suitable model is one based on TeliaSonera being functional [sic] separated. This proposal is inspired by the British model for equal treatment which was introduced recently." (See PTS Proposes Access Separation.)

It looks as if TeliaSonera is not the dullest place to be just now… (See TeliaSonera Dumps Its CEO.)

Elsewhere in Europe, Italian incumbent Telecom Italia (TIM) is also heading down the access network separation road, with the analyst team at Dresdner Kleinwort reporting from a meeting with the carrier that CEO Riccardo Ruggiero is "upbeat on the prospects for a quick agreement with the NRA [national regulatory authority] on functional separation of the fixed network."

In a research note issued today, the Dresdner team states that such separation would help Telecom Italia be more creative with its service marketing. "We are of the view that a favorable result is possible for functional separation of the copper network as TI is presently hamstrung in its ability to bundle services [such as] fixed and mobile in a single bill. Separation would allow more attractive offers to be marketed to the benefit of top line growth."

Ruggiero told the analysts that an agreement with the regulator is due in 2007 "with probable implementation in 2008." They note, though, that Telecom Italia is in the driving seat over the possible move, as "structural separation cannot be enforced and would only be accepted by TI if it was in the company's interests."

On the other side of the world, Telecom New Zealand Ltd. (NYSE: NZT; New Zealand: TEL) faces the challenge of internal separation, which is why, according to The Australian, it is keen on hiring current BT executive Paul Reynolds, a man with experience of how the process works, as its new CEO. (See TNZ puts Reynolds on speed-dial.)

But the operator is concerned it will be asked to invest heavily in new infrastructure only for others to then benefit from the results.

Responding in April this year to initial proposals for structural separation made by the country's Ministry of Economic Development, the carrier's chairman, Wayne Boyd, noted: "Our principal concerns are that the consultation document released last week proposes a very complex form of separation that goes significantly beyond the BT model, and in our opinion fails to address important questions around investment. The proposed form of operational separation creates an independent, but unsustainable Access Network Services unit that has no capability or incentive to invest."

Instead, Boyd is seeking "a structurally separated access network company that would have the ability to earn a commercial rate of return, a simpler separation model so resources can focus on faster delivery of local loop unbundling and Naked DSL, downstream de-regulation to enable Telecom Retail to compete and innovate, and committed broadband network investment from Telecom."

— Ray Le Maistre, International News Editor, Light Reading

Page 1 / 2   >   >>
materialgirl 12/5/2012 | 3:06:47 PM
re: Telcos Consider the Splits By giving visibility to access costs, this type of structural separation makes sense. It makes it harder for service providers to gouge their monopoly "customers" in order to fund expensive white elephants like IPTV.
shumingli 12/5/2012 | 3:06:42 PM
re: Telcos Consider the Splits this is a right direction that will save the cost and advance access network and creat a fair competition of use of asset.

this should be done long time ago.
ipLogic 12/5/2012 | 3:06:41 PM
re: Telcos Consider the Splits The Dinos are going away :-) very good news!

But this addresses competitiveness "only" in Edge and Core architectures. (I don't say it's not good enough, it is definitely way forward)

How can the Access network also become competitive ?

Can they split departments that control the fiber/copper plants, and let 3rd party Operators plug their own equipment in the street cabinets ?

This could further increase competitivness, and see if it is profitable, & faster to build MetroE switches all the way in the street cabinets, or use DSLAMs, or other options. Also we can check technology competitivness like IP in the access vs MPLS PWE vs Ethernet, or small BRAS+Security in street cabinet,

or even "Access Security Providers" with own managed firewalls in the street cabinet into which other 3rd party operators then plug their uplinks and buy the "Security service" for their own subscribers.

Does this makes sense ? or are there any drawbacks ?

i.e. Shuffling things always initiates the sales & development "brain power" to get active and is good start for development of good ideas.

Any thoughts out there on this subject ?

Michael Poole 12/5/2012 | 3:06:41 PM
re: Telcos Consider the Splits Indeed, there were a lot of people who thought this should have been done back in the late 1980s when Telecom NZ was being flogged off, by selling the access network separately. That even included a lot of individuals inside the company! Cynics would say (with an accuracy approaching 100%) that the reason it didn't happen was that it would have reduced the amount of money the government could ask for it. It's quite understandable that Telecom is opposing the regulatory equivalent now, although BT seems to have done quite well despite having its hands chopped off.


paolo.franzoi 12/5/2012 | 3:06:40 PM
re: Telcos Consider the Splits
The problem is that you have made an assumption about street cabinets:

1 - They tend to be full of equipment already, so where are you placing this new equipment.

2 - They are limited in Power, Thermal Capability, and Battery Backup. Retrofitting these capabilities at whose costs?

3 - Multiple Access to a common place needs control. Who pays if Company A' Tech unplugs Company B's equipment by mistake?

Let's use your idea of Metro Ethernet equipment. It takes new copper or more copper than is allocated in the current Outside plant (at least 2 pair are required per customer).

From a Service Provider's standpoint, the equipment that is placed better have a pretty good utilization or it is just not viable. So, in general having a COs base to draw from represents a good idea. Let us say you get a 10% take rate at a cabinet or a CO....

Cabinet - (say 500 lines) - 50 subscribers
CO - (say 25,000 lines) - 2,500 subscribers

Putting lots of small boxes out is a big increase in Opex to provide support and service.

In the US, there is unbundling called UNE-L which is what you are looking for. The problem is the plant that is being rented is designed for the services that the incumbent already is offering. To try to get differentiated offerings, one likely has to rent fiber and place its own copper plant (or fiber plant).

rjmcmahon 12/5/2012 | 3:06:39 PM
re: Telcos Consider the Splits Again we are asking the wrong questions. The goal isn't to make access competitive, that's blind ideology and faith in propaganda and rhetoric. We can and need to set better goals than that. Here is what economist Charles Jacobson has said on the subject a few years back.


1) I think that the goals you mention, motivating investment in fiber optics (particularly, the "last mile" to the consumer which can be enormously capital intensive) while promoting structural separation are worthy goals but may need to be thought through more carefully. In particular, it seems to me that the ultimate goals of separation of application and service providers is to promote technological innovation, and diversity and openness of speech - prevent the content of communications from being monopolized and censored.

2) Separation is of value to the extent that it fulfills these goals, but may not always be the best means of getting there. My own judgment, is that the real prize is not to have three or four choices of competing broadband providers (to use an example from today's internet), but to make sure that the provider that there is has reasonable incentives to provide good service, does not censor content, and does not get in the way of developing innovative applications etc.

3) Care is needed in thinking about competition. Encouraging competition among owners of fiber optics, cables, etc. may actually slow rather than promote the diffusion of more advanced internet technology, investment in fiber etc. The reason is that with many firms potentially competing, firms may not find it worthwhile to make the large initial capital investments required. That is, investing large sums in a fiber optic network maybe worthwhile for a firm if it can have reasonable assurance that it will reap the gains of people using the network. If there is a risk (from the firm's perspective) of other firms building duplicate systems and taking some of the customers, engaging in a price war, the risk of the investment is vastly increased.

4) For public policy, these thoughts suggest that attempts to encourage competition among telephone, cable providers, etc. for provision of internet connectivity, maybe less productive for protecting consumers, encouraging innovation, etc. than accepting possibility of monopoly provision in this domain, or even government investment, under common carrier principles in which the owner of the communications pipeline does not dictate content.
rjmcmahon 12/5/2012 | 3:06:38 PM
re: Telcos Consider the Splits as long as you give Gǣcontrol in single handsGǥ, there is no Gǣstrong enoughGǥ incentive that will fully solve the GǣtemptationGǥ to monopolize, and/or (miss)use the Gǣpower of having full controlGǥ.

Sure there is it's called local control with democratic oversight, i.e. let communities build their own networks and don't hand off the power to somebody else. The problem we face in the US is that we've nationalized too much.

Here is Adam Smith on the subject:

"The abuses which sometimes creep into the local and provincial administration of a local and provincial revenue, how enormous soever they may appear, are in reality, however, almost always very trifling, in comparison of those which commonly take place in the administration and expenditure of the revenue of a great empire. They are, besides, much more easily corrected."

The fallacy is believing that competition (feigned or real) has much to do with the solutions towards modern communications infrastructures. The 1996 Telco Act tried to spur competition and yet 11 years later the net result has been the remonopolization by four telcos, and oversight pushed to a 5 member commission sitting in DC. Tell me exactly how 5 members can address the concerns of 300 million?

So, if prayer to a personal deity isn't slowing the growth of one's cancer maybe try something different and visit an oncologist? Our industries blind faith in "competition" is no different. It isn't working (though the charlatans will adamantly claim differently.)

PS. You may want to read up on natural monopolies.


Utilities are often natural monopolies. In industries with a standardized product and economies of scale, a natural monopoly will often arise. In the case of electricity, all companies provide the same product, the infrastructure required is immense, and the cost of adding one more customer is negligible (up to a point.) Adding one more customer may increase the company's revenue and lowers the average cost of providing for the company's customer base. So long as the average cost of serving customers is decreasing, the larger firm will more efficiently serve the entire customer base.

Historical example

Such a process happened in the water industry in nineteenth century Britain. Up until the mid-nineteenth century, Parliament discouraged municipal involvement in water supply; in 1851, private companies had 60% of the market. Competition amongst the companies in larger industrial towns lowered profit margins, as companies were less able to charge a sufficient price for installation of networks in new areas. In areas with direct competition (with two sets of mains), usually at the edge of companies' territories, profit margins were lowest of all. Such situations resulted in higher costs and lower efficiency, as two networks, neither used to capacity, were used. With a limited number of households that could afford their services, expansion of networks slowed, and many companies were barely profitable. With a lack of water and sanitation claiming thousands of lives in periodic epidemics, municipalisation proceeded rapidly after 1860, and it was municipalities which were able to raise the finance for investment which private companies in many cases could not. A few well-run private companies which worked together with their local towns and cities (gaining legal monopolies and thereby the financial security to invest as required) did survive..
paolo.franzoi 12/5/2012 | 3:06:38 PM
re: Telcos Consider the Splits

So, you are okay if the equipment you are personally connected to is "amortized" out of the cabinet and replaced by some other product you do not connect to?

I am surprised that this would be okay for you.

Fiber is only a default for large enterprises. Most customers are not large enterprises.

There are many more things you have missed but that is a couple of biggies.

ipLogic 12/5/2012 | 3:06:38 PM
re: Telcos Consider the Splits seven,

i know i'm missing stuff, but still looking for the pros and cons, and how "todays technological environment" might influence this.

when I moved in my appartment I just had a copper cable not connected anywhere, then the telco guy came connected the wires in the basement, and afterwards pluged us in the nearest DSLAM.

in (lets say) 4-5 years the DSLAM investment of the telco should be payed off, and they could (should) remove it and let the street cabinet be shared among multiple operators, each one with own DSLAM/MetroEswitch/whatever, and I can choose among them.

Its like:
1. Construction company builds huge shopping mall Building
2. "Shopping center company" operates the building (cleaning, power, etc)
3. "Clothing/Distribution stores" rent space and infrastructure-elements in the mall, and sell goods
4. Fashion houses produce the goods

1. "Construction company" builds Housing+Fiber FTTH
2. "Medium-provider" operates the Fibers, Street Cabinet (L1)
3. "Network/Access-provider" operates access-&-network technology with goal to sell/distribute connectivity (L2-L3)
4. "Content providers" produce content (L7) and let the Network Operator distribute it to subscribers

Each one does what they are best at, and the similar compete against each other.

Finally customers gets the best of breed of each part of the puzzle. And noone can monopolize in some way.

It somehow makes sense to me.

ipLogic 12/5/2012 | 3:06:38 PM
re: Telcos Consider the Splits seven, rjmcmahon, thanks for your thoughts !

in summary:
- first-mile communication-medium (especially fiber) is (default) part of Gǣbasic-living-infrastructureGǥ, so it should not be seen (any more) as huge investment; just like we have water and electricity everywhere and no one considers building the plumbing or electricity network as Gǣrisky investmentGǥ;
(eventually the state might be of assistance in the parts where Living-infrastructure was not enriched with communication-infrastructure from the beginning)

- the problem is the management of many small boxes is defenitely Opex factor, at least until we find Gǣeasy manageableGǥ and reliable small devices;
- already installed equipment in street cabinets should be given chance to be amortizied and then the space freed-up for the new competitive-access model;
- equipment utilization risk and cost is always there, and is part of every operational model, no matter who owns the parts;
- since fiber is becoming GǣtypicalGǥ (default) underlaying L1 medium for any L2/L3 technology, so this makes Gǣservice dependenceGǥ less and less a show stopper.

the first thing I can think off is the Wireless space, where the "Frequencies Medium" is rented and shared by different SP's. So if we look at the fiber/copper plant as "just another Medium" than it might be also shared and rented.

so if we define GǣMedium ProvidersGǥ (just GǣFiber guysGǥ/GǥDigging guysGǥ without any technological background) they might find their own *Profitable* business model taking care of the Fiber plant and Street cabinets (including power). So this GǣMedium ProvidersGǥ take care of Power, Thermal Capability, and Battery Backup in the Street cabinet, and *Rent* this in their package.

The GǣMedium ProviderGǥ does have his own business case of building the Gǣmedium(fiber)-plantsGǥ, since there is almost no family/house that does not use communications of some type. So this investment in Gǣcommunication mediumGǥ is at very low risk of not being utilized.

On top of this GǣMedium ProviderGǥ then come the GǣAccess ProvidersGǥ which put their own equipment/technology, and Rent space-&-power in the street cabinet, and rent the medium (fibers/copper link) to their subsribers.

This might increase Gǣecological competitivnessGǥ by pressuring vendors for low-power, and utilization efficient devices (like stackable devices)
(Company A and B in wireless also have wireless stations on common places, but they still find a way to secure access to the equipment. so this is not so big show-stopper)

Utilization of access equipment is always at GǣsomeoneGsGǥ cost (i.e. using 50 out of 500 lines is always Gǣbeing paidGǥ by someone) So this is not weakness of this model of GǣMedium-Access-separationGǥ, but is weakness of *any* operational model.

GǣFrequencies MediumGǥ is Gǣfree of chargeGǥ i.e. no cost of build-out;
So what about the cost/investment of build-out the GǣFiber mediumGǥ
well nobody buys a house without Street, Power, Electricity, Gas, Water, **and** Optics (or at least Copper).

So if years ago the GǣLiving InfrastructureGǥ was built without GǣCommunication InfrastructureGǥ, this is not the case anymore. GǣLiving+CommunicationGǥ infrastructure come together i.e. so we donGt have (anymore) this kind of Gǣhuge Initial InvestmentGǥ in communication medium.

>> rjmcmahon >> 2) ...to make sure that the provider that there is has reasonable incentives to provide good service, does not censor content, and does not get in the way of developing innovative applications etc.

as long as you give Gǣcontrol in single handsGǥ, there is no Gǣstrong enoughGǥ incentive that will fully solve the GǣtemptationGǥ to monopolize, and/or (miss)use the Gǣpower of having full controlGǥ.

itGs like saying Gǣwe donGt need democracy, because the emperor will promise to behave niceGǥ :-)
there are examples of Gǣnice behaving emperorsGǥ but as well a lot of opposite examples.

Page 1 / 2   >   >>
Sign In