Back to the Future on Unbundling

11:50 AM -- After decades in telecom, Keith Willetts isn't pulling his punches. In his new book, Unzipping the Digital World, the chairman of the TM Forum takes square aim at two groups whom he views as obstacles to the success of telecom service providers in the digital era: regulators and financial analysts. (See Willetts Unzips Telecom Future.)

Dig a little deeper into Willetts's ideas on regulation, however, and it's apparent he's not championing any ideas that U.S. carriers, such as AT&T, Verizon and Comcast, can use to shake their fists at the Federal Communications Commission (FCC) .

Willetts does believe that, in general, regulators are living in the past, remaining too fixated on breaking up the old monopolies. But in an interview in Dublin last week, during Management World 2012, he made it clear he principally is railing against those regulators who advocate building multiple physical networks. In our LRTV video interview with Willetts, he got talking a lot about enabling new services, in fact:

Anyway, building multiple physical networks makes no economic sense, Willetts says, given the overall costs and the time required to earn a return on that investment. But Willetts does see value in competition at the services level. He favors a concept that has been kicked around -- unsuccessfully -- for years now: forced unbundling of the existing physical networks.

The U.K. attempted such a thing with BT, forcing the company to wholesale its local loop, but failed to require ubiquitous broadband services, leaving those who live in less populated areas -- including Willetts himself -- bandwidth starved. His best connection is 768 Kbit/s.

"We screwed up unbundling in the U.K. because OpenReach has no universal service obligation," he says.

Let's be more like ... Malaysia?
By contrast, in places such as Malaysia, a public-private partnership built an FTTH network in a way that enabled government funding of the basic infrastructure that private service providers now pay to use, Willetts says.

"From a comms perspective, Western Europe and the U.S. are seriously disadvantaged -- not having a broadband comms infrastructure is like not having airports or roads," he says. "Seventy-five percent of the world's FTTH is in southeast Asia."

In the U.S., the very notion of unbundling a wholesale network has run afoul of Willetts's other main punching bag: Wall Street. Since the network builders in the U.S. are publicly traded companies, they have investors to satisfy and those pesky financial analysts to impress. The U.S investment community was less than impressed back in the early 2000s when Verizon started talking about building fiber to the home in the densely populated portions of its service footprint.

If the idea had been to immediately allow others to sell their services over that network -- say, the local cable companies and some CLECs -- the investment community would have been even less impressed and FiOS might not have happened. Even the FCC, which under the 1996 Telecom Act was instructed to force unbundling of the local loop, drew the line at including new broadband networks, believing that would stifle investment.

And that, says Willetts, may be where everyone is making a mistake.

He has a point -- Netflix Inc. (Nasdaq: NFLX) could hardly be in Washington complaining about unfair treatment if it was simply allowed to buy access to an open local loop. Ironically, having to pay for that access might destroy its business plan and force it to increase prices. Netflix counts on having free access to broadband pipes in order to stream its content at low cost, and thus attract a large volume of consumers.

But the more solid point that Willetts makes is that the investment community, with its rock-hard fixation on next quarter's earnings, is doing big telecom no favor. And he also takes off after regulators for using spectrum as a means of raising incredible sums of cash for financially strapped governments, even though that process immediately puts the spectrum owner in a deep hole financially, as Europe learned the hard way with 3G.

These and other points in the book are well-taken, but it's hard to see anything changing. Neither Wall Street nor regulators have much sympathy or concern for large telecom operators, and are unlikely to re-order their worlds to match the needs telecom's needs. It is likely to remain the other way around for some time to come.

— Carol Wilson, Chief Editor, Events, Light Reading

paolo.franzoi 12/5/2012 | 5:31:47 PM
re: Back to the Future on Unbundling




You mixed metaphors quite deftly there and I think that you have many valid points, but you need to clarify this.


You go from the commentary about Local Loop Unbundling, where a competitor rented a line that was owned by an Incumbent to complaints about Bandwidth Caps by Netflix.  There is an analog there but they are very different.  Netflix does not want to rent a line and become the customer's ISP.   And I think this is where the problem and confusion around all this starts.


There are really 5 entities here:  Content Provider, Internet Core Provider, Internet Service Provider, Access Service Provider and Customer.  In some cases, several of those are within the same company (the large ISPs are generally Tier 1 ISPs and have direct access to "The Internet").


Where I see the problem different than most is that in the olden days (90s and early 00s) their was the notion of ASP and ISP truly being two entities.  The telcos and cable cos killed this in general or have worked very hard for this to be dead.  Why the cable cos can get away with it 100% and the telcos can not is beyond me.  To do this, they made the ISP portion of the business "free" and put all the cost to the customer in the ASP portion of the business.


It turns out that being an ISP does not cost 0 and I do not think the problem is with the ASPs but with the ISPs.  An ASP does not care about bandwidth caps because to him or her each endpoint has bandwidth that is unique.  The problem comes in where they are heavily aggregated - at the ISP.  The ASP has to build networks to sustain large bursts of traffic at maximum rate (hey I want to stream an HD movie).  It is the ISP that doesn't want 10K HD movies being streamed.


Now I use this as an example to try to illustrate a point.  These caps are applied by the ISP business - which nowadays is not a great business to be in.  Imagine if you did structural separation FOR ALL NETWORKS and all ISPs with their 0 revenue poof went out of the residential ISP business.


So, first I think it is a bit more complicated and second I think we need to be clear about the issues.




cnwedit 12/5/2012 | 5:31:46 PM
re: Back to the Future on Unbundling

Actually, I'm not mixing metaphors. Your definition of how the industry breaks down is accurate but dated, in my opinion. I realize that there is a difference between providing the physicl network and providing Internet access but I think the industry is rapidly evolving away from the traditional definitions because of how services and applications are being offered OTT today.

I don't consider Netflix a "content provider" - they are a service provider. They aggregate content created by others and deliver it over a network as a subscription service. They are paying for that content and they&nbsp;want that network to be sevice-neutral, which it isn't today, and that's not just a matter of bandwidth caps, but also quality of service, which includes things such as latency. QoS will become much more important as the network incorporates more wireless endpoints.

Unbundling the network doesn't have to be&nbsp;limited to selling T-1s or wavelengths - there&nbsp;could be&nbsp;a reasonable business model around creating network service packages, if we refuse to be bound by the old ways of looking at things.

Certainly the old way of unbundling the access network was much more limited, and ISPs would be the service providers. But that horse left the barn years ago.



paolo.franzoi 12/5/2012 | 5:31:42 PM
re: Back to the Future on Unbundling


The purpose of your unbundling is to get invenstment in access correct? &nbsp;To get more FTTH.

I agree with the government part but the rest of it is rather complete bunk. &nbsp;There is no reason to believe that more dollars will cause anything more than even greater profits in networks. &nbsp;It is NOT that there is no business case for FTTH. &nbsp;It is that there are BETTER business cases in other things. &nbsp;That is the beauty of the incumbent Service Provider position. &nbsp;Unless they are completely daft they will make money. &nbsp;

There have been 2 reasons for FTTH investment:

Competition (NTT and Verizon)

Government Intervention (KT and I guess Malaysia)

No amount of any other source of money will do it. &nbsp;Netflix works just fine on U-verse. &nbsp;Heck it works fine on regular DSL. &nbsp;The issue for Netflix is not a broken access technology. &nbsp;It is people trying to get money out of their ISP business again (instead of their ASP business).

Look at the US incumbent telcos and with the exception of Fairpoint (which really over reached) they are ALL awash in profits. &nbsp;Unless you create a Universal Service Obligation AT&amp;T will not go out into its rural areas and deploy U-verse, FTTC (which by the way has the same regulatory relief as FTTH), or FTTH. &nbsp;They just see better ROI in spending their money on Wireless. &nbsp;If they lose every subscriber in Laytonville, CA they don't care. &nbsp;The base problem is nobody wants to spend the money to take those customers away.

All of this leads back to where you have mixed things up. &nbsp;Giving an incumbent a wad of cash will NOT cause them to invest in their access networks. &nbsp;If they wanted to they could do it today AND it has a positive ROI (albeit with a break even further out than they would really like). &nbsp;There are only two ways to get them to invest and those are both forceful means: Competiton and Regulation.



Sign In