Skip to main content

Telcos must double capital investment to succeed with IPTV deployment

A new white paper from telecom consulting firm Detecon sheds light on the challenges telecommunications companies face as they begin rolling out IPTV service in the United States.

Not only must telcos double their capital investment and overcome unresolved questions about how set-top boxes will be deployed in the home, they must also take up this IPTV business under do-or-die circumstances as cable MSOs whittle away at their core business with triple play services, says paper author Dr. Patrick Pfeffer, Chief Network Architect at Detecon.

IPTV Update thought it was a good time to interview Pfeffer to learn more about the conclusions he’s reached in “IPTV: Technology and Development Predictions.”

IPTV Update: MSOs adding voice to existing video and data services creates a situation you describe as a last chance for RBOCs in which adding video is critical. Will Verizon FiOS and AT&T efforts get it done in time?

Patrick Pfeffer: If you look at the history of telcos and video it is one of less than stellar performance. In the mid ‘90s telcos attempted to go into video. The driver was revenue growth. They failed. Today, some are again attempting the same journey towards video but the driver is fear. Their core business, which is voice, has been declining over the last few years. Today telcos have 150 million access lines. By 2010, the number of access lines will be down to 130 million. Initially, telcos were losing voice lines to wireless carriers. Today, the MSOs — after much trial and error with VoIP — are also offering voice.

For example, today Time Warner has more than 1 million VoIP subscribers. The service is called Digital Phone. That’s more than Vonage, even though Vonage has invested millions in marketing campaigns.

Telcos have no choice. They have to offer video.

Verizon, AT&T and BellSouth are taking advantage of a 2004 FCC ruling. For the last 10 years — since the 1996 Telecommunications Act — the policy was to promote competition in the local loop by unbundling. In 2004, the FCC reversed its position, setting forward that the local loop no longer needed to be unbundled. If fiber is brought to the home there is no unbundling requirement. This architecture, called FTTH or fiber to the home, has been chosen by Verizon. If fiber is pushed up to within 300ft of the end user, the terminated copper pair does not need to be unbundled. This architecture, called FTTC or fiber to the curb, has been chosen by BellSouth. Finally, if the copper pair only supports data or video then there is no unbundling requirement. This architecture, called FTTN or fiber to the node, has been chosen by AT&T.

That meant a lot of investment by the TriBOC (AT&T, BellSouth and Verizon).

The architecture options lead to different operational models. Verizon calls its deployment FiOS and it is characterized as an FTTP architecture where “P” stands for premise and encompasses both residential and commercial buildings. The cost to reach a premise ranges between $1200 and $1500. Verizon is installing fiber all the way to the end users, a technician cuts the copper line, and there is no way back. With FiOS, Verizon is operating like a cable company. It is not really IPTV. Verizon must acquire a franchise in each city it plans to offer video. It is costly, time-consuming and doesn’t scale.

On the other hand, AT&T has decided to go with IPTV. AT&T is further mining its copper plant, and it hopes to connect a home for less than $500. AT&T believes IPTV can scale, and the proof is in the pudding. AT&T believes it doesn’t need a franchise to offer video services since they are carried over data.

The FCC needs to provide a national broadband policy that clearly defines the rules of engagement. The lack of clarity creates uncertainty and impedes deployment progress.

Another major difference between Verizon and AT&T is the choice of equipment provider. Verizon is going with a best of breed approach, using a mix of hardware and software from a wide variety of vendors. AT&T is using Microsoft, Microsoft and Alcatel. Microsoft’s proposal is very appealing because it offers one-stop shopping. Its integrated solution provides, at least in theory, the hope that integration issues that have plagued previous telco video effort can be better managed.

IPTVU: In the white paper, you said ADSL isn’t up to supporting IPTV deployment, and that RBOCs should double their capital expenditure. Can you explain?

PP: The capital expenditure is related to the network, and I think it is important to compare that investment with other telecommunications investments. If you look at what the cable industry has invested since the 1996 Telecommunications Act, you are looking at $100 billion. If you spread that out across all cable subscribers, it represents $1400 per subscriber.

Similarly, look at Cingular. It invested $1740 per subscriber when it acquired AT&T Wireless. The average revenue per user (ARPU) for a cable user is about $70, the ARPU for a postpaid wireless subscriber is about $55, and the ARPU for a household connected to IPTV network is about $150 when it includes triple play (voice, video and high speed data). I believe such revenue potential justifies an investment of $2000 per home. As we look at the 110 million homes in the United States, I would like to see $200 billion committed to the necessary network upgrade. It will take many years, but it has to happen.

Telecommunications is very capital intensive with rapid technology changes. Wall Street does not understand this and continues to monitor the telcos on a quarterly basis. Telecommunications investment is a solid investment, but does not bring instant gratification.

IPTVU: You find that Microsoft TV middeware is becoming a de facto standard. Will that give Microsoft an inordinate sway over the future direction of IPTV deployment, and is it good or bad to have Microsoft as a key driver in the IPTV space?

PP: Microsoft has a history of succeeding when it is committed to an industry, even when it’s late to that industry. Microsoft has invested $20 billion in video, cable and let’s say TV. Today they have a solution that is appealing for all large telcos. Telcos can turn to Microsoft, and Microsoft will be one-stop shopping. Wall Street is also comforted that the telcos are collaborating with Microsoft.

The approach is not without risk. Will Microsoft be able to deliver? And when? Additionally, there is no back-up solution. Now you have the sempiternal conundrum: Is Microsoft’s dominance on the desktop speeding up or slowing down progress? The same can be asked for IPTV. You have a company that can invest billions of dollars, offers an economy of scale and a value add. Today, Microsoft is good for IPTV and it is jump-starting the market.

IPTVU: Can you discuss the last 100m and last 10 centimeter problems telcos have and possible solutions?

PP: The last 100 meter issue is, “How do you wire the home?” Today in the United States, you have 2.8 TVs per home. If you provide IPTV, you are going to need a set top box for every TV, and you will need to bring the wire to every TV. That is a major problem. All of the pioneers in IPTV had to spend four to five hours per home to bring that wire to the set-top boxes. A solution must be found; otherwise it kills any business plan.

Verizon or the MSOs can avoid this problem by offering on the same infrastructure up to 80 analog channels that can be viewed without a STB. It is also worth noting that DBS providers have for many years dealt successfully with multiple STBs but at a high subsidy cost.

In terms of in-house wiring, there are a number of solutions, such as multimedia over cable that reuses the coax in the home to bring IPTV channel. There is another solution based on a new version (802.11n) of the WiFi standard with higher bandwidth and the quality of service mandatory to offer video services.

The last 10-centimeter issue relates to the set-top box. The set-top-box is still the most difficult element in the IPTV proposal both from an engineering standpoint and a business standpoint. It is like a new computer, but not very stable. It’s difficult for telcos to stabilize their platform. The other issue is cost. Deployment costs scale linearly, which is not good.

New STB designs based on Broadcom, Sigma, LSI Logic and Connexant SoC (System on a Chip) should bring the cost down. I expect to see reference designs from these companies based on a couple of chipsets. When we reach that level of integration, Asian-based companies will be able to manufacture set-top-boxes with a build of materials in the $30 range with an additional $10 to $15 for the mechanic. This is really something the telcos and their partners are going to have to work on. It will require large volume. It is a make or break for IPTV.

The last issue with the last 100 meters and the last 10 centimeters has to do with installation. Truck rolls must be avoided. It took 10 years for telcos to offer ADSL systems that can be self-installed by users. The same thing has to happen for IPTV and sooner than later. Otherwise, IPTV just doesn’t work.

To read an executive summary of the paper, visit

Tell Us What You Think! IPTVU invites response from our readers. Please submit your comments to We'll follow up with your comments in an upcoming issue.

Back to the top