On October 1, 1906, H.M.S. Dreadnaught slipped into the English Channel, one year and one day from the laying of her keel. Over the next few days, she conducted high-speed runs in excess of 21 knots displaying the power of her steam turbine engines, a first for a warship of her size (18,000 tons). Her ten 12-inch guns could engage targets at 10,000 yards. She was fast enough to outrun any ship her size and she was powerful enough to outgun any other ship of her class. Jane’s declared she was the equivalent of two or three of battleships of the designs that had preceded her. Dreadnaught was a lethal and brilliant combination of striking power, speed, reliability and armor that could choose to fight when, where and who she desired. In 1907 Dreadnaught became the flagship of English Home Fleet and assumed the responsibility of the guardian of Pax Britannia at the very zenith of the Empire and only six years after the death of Queen Victoria.
When details of her design became public, it was apparent that this single ship hailed as a triumph by her designers, was in fact a colossal blunder in the eyes of many. It took time for the full impact of Dreadnaught to be realized. Her combination of speed and striking power turned every ship in every navy, in every port of the world obsolete in 31 months. Critics called Dreadnaught:
“[A] piece of wanton and profligate ostentation” that “…morally scraped and labeled obsolete [the entire British Fleet] at the moment when it was at the zenith of its efficiency and not equal to two, but practically to all the other navies of the world combined.”
David Lloyd George said:
“We said, ‘Let there be dreadnaughts’ what for? We did not require them. Nobody was building them and if anyone had started building them, we, with our greater shipbuilding resources, could have built them faster then any country in the world.”
In the end, the critics were wrong and the pioneers were correct. Reginald Bacon, Dreadnaught’s first captain said to the critics:
Knowing as we did that Dreadnaught was the best type to build, should we knowingly have built the second-best type of ship? What would have been the verdict of the country [i.e.England] if Germany…had built the Dreadnaught? What would have been the verdict of the country if a subsequent inquiry had elicited the fact that those responsible at the Admiralty for the safety of the nation had deliberately recommended the building of second-class ships?
The history of this era is beautifully composed by Robert Massie in two books: Dreadnaught and Castles of Steel. I generously borrowed from Massie’s first book Dreadnaught (pages 468-489) to compose the first three paragraphs. If you enjoy naval history, The Grand Fleet and Battlecruisers are two cherished books in my library that complement Massie’s books.
Friday, March 2nd 2007:
It is Friday afternoon. The market just closed. I am looking forward to nice weekend with the family with plans to attend Chinese New Year festivities in Chinatown in Boston. All is well in my world as I review various blogs and media sites for end of the day news and reports. First I see this post. Then I see this post. The two posts motivate me to compose this post on and off over the next week.
Wednesday March 7th 2007:
As I am in the process of draft elements of this post, I read this article in the WSJ. I click over to see what Lightreading.com is blogging about and read this article on FCC video rules.
To write a meaningful post on FiOS that does reiterate the observation of the obvious requires an understanding of the past and a hypothesis of the effect that FiOS will have on the long term state of expectations within markets. Here is what I am not going to write about:
It costs too much! We do not know what the future will look like when it is done! Why would anyone need all that speed? Verizon should buy a satellite TV company Why would anyone switch to Verizon? Cable is so far ahead, Verizon will never catch up!
Here is a quote from the 24/7 Wall St. article on Verizon and FiOS:
Verizon plans to have its service available for 23 million homes by 2010. Fortune recently quoted an analyst who is concern about the Verizon plan: "We view this as a multiyear issue," says Citigroup telecom analyst Michael Rollins, who predicts that the pain will carry into 2008. "The market needs to be braced for a longer period of dilution and higher-than-expected costs from this FiOS build." Not a very nice thing to say. The business magazine points out another negative: "FiOS's awesome speed presents a potential long-term challenge: It may encourage consumers to download movies, TV shows, and other interactive services directly from the Web, bypassing Verizon's video offerings entirely." Cable is light years ahead in terms of offering the dreaded "triple play" of TV, broadband, and phone service (VoIP) to consumers. Why a customer would switch to the Verizon fiber product is a mystery. Perhaps Verizon should stop building and use the money to buy a satellite TV company instead. There are two that may be available.
Assumption #1: Yes, FiOS is a massive capital investment ($23B is the often quoted capital commitment). Yes, many investors would prefer to have Verizon invest the cash in their dividend, debt repayment or stock buyback. Yes, it is a multi-year commitment of capital because Verizon is replacing an aging installed base of copper that was deployed over a 30-50 year period. Replacing that amount of fiber deployed over decades is clearly a multi-year project.
Here is a link to a recent paper by David Levinson and Andrew Odlyzko. Although I was recently accused of being a blogger long on time and short substance that does not do real analysis, by a person who posts on message boards nonetheless, I do make an effort to read a large number of books and papers across a broad set of subject matter. If you do not want to read this paper, I will extract a few points from the paper that I will use to frame concepts in this post. I will start with an extraction of four economic terms found at the start of the paper:
Excludable: Provider can prevent others from using Non-Excludable: Provider cannot prevent a user from using a good or service Rivalrous: Users can prevent other users from consumption Non-Rival: Users cannot prevent other users from consumption
Assumption #2: Users prefer flat rate pricing over metered pricing for three reasons: (1) insurance effect, (2) overestimation of usage and (3) potential hassle factor or peace of mind. [see Levinson/Odlyzko (.pdf) pages 6-7]
Assumption #3: Flat rate plans tend to increase usage of a service 50 to 200%. Interesting summary of flat rate dial-up internet access on page 16 that shows how flat rate local calling combined with a flat rate access charge tripled internet usage. [see Levinson/Odlyzko page 7]
Assumption #4: Many technologies under development are focused on providing fine scale charging or metering, while flat rate based usage plans are spreading. [see Levinson/Odlyzko page 15]
Assumption #5: QoS based architectures and methods have consistently failed economically. It has always proven less expensive to add capacity than to implement and administer complex prioritization methods. [see Levinson/Odlyzko page 15]
Assumption #6: Voice is still the dominant revenue engine of the telecommunications industry. [see Levinson/Odlyzko page 17]
Assumption #7: The cost of collecting charges on a transaction based usage plan by a service provider is often too costly and the increased benefits are too marginal that the outcome is that metered billing plans with QoS functions are too costly to bother. [see Levinson/Odlyzko page 18]
Assumption #8: Verizon’s FTTP program called FiOS is replicating a broadcast, QAM based, RF architecture for the delivery of video that is virtually identical to current cable architectures. Most other telecom centric service providers are deploying a switched video delivery system that utilizes the embedded base of copper and DSL infrastructure. There are challenges with the IPTV such as: latency, signal degradation, bandwidth limitations, ability to scale to meet increased HD content, content licensing and protection methods and it is still unproven and will take time to sort out. It will work, but it will take time. Many investors who criticize FiOS for taking years should ask how long it will take to perfect IPTV and turn it into a service that competitively takes market share in a zero sum game from other providers of video and other services.
Assumption #9: Service providers are all moving in the direction of a common set of services (i.e. video, data, voice, and mobile) from different heritages and strengths. A data rate war looms. Shooting has already started. I have posted on this subject in the past. Here are the links to related articles: Link-1, Link-2, Link-3 and Link-4.
Assumption #10: Cable infrastructure has a good position for the near-term, but it is only a matter of time before there is another spectrum upgrade as more HD content becomes available and broadcast groupings continue to narrow to provide consumer choice and increased revenues from advertising in narrower groups.
Assumption #11: An email exchange with Scott Berry last week, who has posted comments on this blog and was a partial subject for the follow-up post on internet video, he started me thinking about the value of the annual revenue for mobile, entertainment and communications on a per unit basis. I then began to think of a mapping this value into Maslow, but this is for another day and another post. Simplified Mobile-Entertainment-Communications [MEC] Revenue per Unit [MECRPU] hypothesis chart is shown below. Assuming there are minority, but extreme users who spend no capital or large capital expenditures on MEC, this model is designed to reflect three types of general user values. The question is which suppliers are accessing components of this revenue? The total MEC dollars per user must be finite. MEC suppliers who offer more and better MEC options are doing so to take MEC market share from competitors. New MEC goods and services do not imply that the MEC market is larger – but rather that a competitor is losing MEC revenues and market share has shifted to a new supplier. The MEC market is growing, but I would hypothesis that new MEC options take market share from legacy MEC offerings, rather then create wholly new MEC markets. This is an example of a non-excludable and non-rival service. When we examine how services are offered in MEC markets, we often find walled gardens and fixed length contracts which are methods employed by service providers to create excludable services.
Assumption #12: As a corollary to the assumption that a measurable Mobile-Entertainment Communications Revenue [MEC] per Unit value exists, then an MEC Time per Unit [MECTPU] value must also exist. These two values must affect each other. End-users only have a finite amount of capital and time to spend on MEC. I would not go as far as to say that MEC markets are fully deregulated, but technology does create market structures in which service providers face the challenge of other MEC providers offering non-excludable services and end-users can act as rivalrous forces. The underlying premise of the Akamai (AKAM) business model is the distribution of content on servers that can find paths round rivalrous areas created by users accessing content. As with the questionable potential of a maximum MECRPU value, there must be a finite amount of time that users spend on MEC. If the actual time spent on MEC per user is finite, then the metric that needs to be measured is how is this time shifting or in other terms how is market share shifting between MEC providers? Less time spent watching TV means more time on the internet? As with MEC capital, this is really a question of stealing time from other MEC providers and with time comes revenue. Here are some questions for another day and another post: Are there net new MEC revenue streams being created? Is it a question of market share exchange in terms of revenue and time that MEC providers are really competing for?
Current State of FiOS
I was not in the room when the final decision was made to go forward with the FiOS initiative, but I suppose that if Jackie Fisher was in the room, with no knowledge of telecom and the internet, he would have found similarity and felt at ease with the decision to undertake FiOS.
Building FiOS was the correct decision because it was the best choice. Anything less then replacing an aging base of copper with fiber optics was to choose a lesser solution. It is true that FiOS is a great undertaking, but it is a necessary choice. Andy Grove once said:
“…[The] boom was healthy too, even with its excesses. Because what this incredible valuation craze did was draw untold sums of billions of dollars into building the Internet infrastructure. The hundreds of billions of dollars that got invested in telecommunications, for example. You know, when the information highway was the craze, the question I would ask [then-Bell Atlantic CEO] Ray Smith and [then-TCI chair] John Malone was, Who the hell is going to spend the billions of dollars it will take to build this thing out? You guys? The federal government? It's not going to happen. And no one could give me an answer as to who was going to pay.
Well, it turns out that the answer was the investing public, who rabidly ran and shoved the money into the hands of the infrastructure builders. It is probably true that the infrastructure would have gotten built anyway. But instead of it happening over 15 years, it happened over 5, because of the gold rush mentality and all these investors trying to get in on it. So the boom accelerated the deployment of the infrastructure, and I'm talking about the Amazons of the world as much as the JDS Uniphases. Amazon's database is a kind of infrastructure - commerce-related infrastructure. When [Merrill Lynch analyst] Henry Blodget projected Amazon would go to $400 and the investing public rushed in, they were funding the deployment of Amazon's infrastructure, which is part of the totality of the Internet infrastructure. And all I can think is, how would this all have happened any other way?” [see Andy Grove, Wired 9.06, June 2001].
What Andy did not realize in 2001, was that we are far short of infrastructure required to enable the vision of a New Economy with lower cost economic transaction models. The 1990s telecom boom did many things, but it did not solve the problem that was the impetus for break up of AT&T: copper local loops. The revolution against AT&T started with Microwave Communication Incorporated, or MCI. MCI used microwave technology to build private networks that bypassed the legacy copper local loops used by AT&T’s (T) local telephony divisions.
Over time, what MCI gained that was significant was the right to offer long distance telephone services. Initially, MCI had limited market coverage and the service was not to the same quality level as AT&T’s, but in the fullness of time MCI built a business by offering discount long distance services. Note assumption #5 is applicable again. MCI raised debt and continued to build an improved network. By the early 1980s, MCI was becoming a global company and a worthy competitor to AT&T. The litigation between AT&T and MCI is what led to the eventual ruling by Judge Harold H. Greene that AT&T was a monopoly. To the settle the case, AT&T leaders decided to break up the company.
A breakup would satisfy the government complaint that AT&T was a monopoly and it would provide a solution to the problem of the legacy local copper loops. The legacy of the copper local loops had been considered by the executive leadership team a plague on AT&T since the end of the Second World War. AT&T had lobbied against microwave technology because it could be used to rapidly build an inexpensive alternative to AT&T’s networks – thus destroying their sunk cost investment in copper local loops.
By breaking up the company, AT&T’s executives found a solution to the legacy copper local loop, their perceived Sword of Damocles, as well as a solution to their legal dispute with the U.S. Government. Fast forward to 2001 and despite the rise of the internet, wireless, cable, CLECs, the main challenge for the RBOCs/ILECs has remained constant over fifty years: copper local loops. Why is Verizon funding FiOS? They are building FiOS for a few reasons that are not often published and all of them are related to copper local loops.
(1) They need to reduce churn for voice services (Assumption #6). Start by thinking of Level(3) and Qwest. What is the real difference between a company like Level(3) and Qwest? Qwest has 13.8M local access lines (i.e. copper local loops) that pay a monthly fee. Joe Nacchio may have been known for waging an LDD rate war in the 1980s at AT&T, but he did not need divine guidance to make the decision to buy US West when he could. It is a good move anytime a new entrant can buy an installed base of tens of millions of customers who provide monthly income. The wholesale only backbone service providers can go find all the content customers they want, it will never add up to having millions of customers who pay $40-80 a month for local voice service. Wholesale fiber providers will always have to deal with Assumption #5 if they think capacity will become scarce, thus turning bandwidth into a valuable commodity and fiber to gold. If having a backbone was such a great business, why did AT&T and MCI sell out and Sprint (S) merge with Nextel? This leads back to FiOS.
Verizon is funding FiOS to reduce churn of services starting with voice. Churn occurs because cable service providers are offering a triple play. Churn occurs because mobile networks have high enough quality levels and low enough rates in metro areas for subscribers to discontinue local access. Churn occurs because cable service providers offer cheap voice, cool TV channels and high speed internet. Users like flat rate pricing and a single provider to pay each month.
(2) Assuming the historical tendency for users to prefer flat rate pricing (Assumption #2), service providers must differentiate on speed, content and breadth of services offering. FiOS will enable Verizon to have a vast reservoir of capacity to fight a bandwidth rate war. Rate in term of speed – not cost. Assume cost is a flat rate across competitors; FiOS will enable Verizon to offer vastly more bandwidth at equal cost. FiOS should enable Verizon to reduce OPEX by moving to a simpler billing system and not bringing forward legacy billing systems and adding fine scale architectures such as IMS. Note this could be unfavorable for Telcordia.
(3) Flat rate plans coupled with increased end-user usage, places Verizon in a position to couple content and applications with their network. Yes, the speed and capacity will enable third parties to provide non-excludable content and applications that Verizon cannot monetize, but Verizon is a large company with ability to fund the development of content and applications or acquire content entities. The other benefit that Verizon will realize long term with their FTTP architecture is they will not be forced into implementing complex, rate prioritization methods to support one type of content or service over another. This assumes that HD content will continue to increase in the form of broadcast media, movies and gaming.
(4) When the FiOS infrastructure reaches the ~50% deployment level, Verizon will be in a position unrivaled in the market. Verizon expects to achieve this level by year end 2010. If they do, they will have 23M homes passed, a large mobile offering from their wireless business unit, an equivalent cable network architecture deployed and the highest quality data service available. This is the point at which I think they can start to inflict pain on their competitors. Pain is the acquisition of market share in a zero sum game.
Here are four charts on FiOS. The first is taken from Verizon’s last quarterly report. The second is a deployment model I created based on limited historical FiOS reports from Verizon, deployment goals stated by the company and anecdotal data points I received from various people. (Note: This is not real analysis as my detractors have pointed out; I produced this model sitting on my couch watching a show on the secret archeology of Rome on the History International channel on Saturday afternoon). The third and fourth are charts derived from the FiOS deployment model.
Dreadnaught’s effect on the state of long term expectations
There is perhaps, a historic parallel between the effect that Dreadnaught had on world in 1907 to the effect of Verizon’s FTTP in 2007. When Dreadnaught was fitting out, England's primary European rival Germany was deeply committed to an alternative strategy. Germany had launched the Deutschland class battleship in November 1904. The Deutschland was the first of five ships of her class with two more ships following in 1905. All three were obsolete from the moment they joined the fleet because Jackie Fischer and England's Naval Board had made a bold decision that disrupted the long term state of expectations in the naval arms race. Dreadnaught created a serious conundrum for Germany. German battleships were limited to the transit restrictions of the Kiel Canal, which was the passage for the German High Seas fleet to the North Sea from the Baltic.
To build a ship that matched or exceed the capabilities of Dreadnaught required the widening and deepening of the 61 mile long Kiel Canal. Germany had the choice of building inferior ships or trenching a new canal. In 1907, after a year of indecisiveness in which they stopped all ship construction, they began work to widen the canal. It would take years and millions of marks to fund the canal project. 1907 also was the year that Germany began construction of the first ships to match Dreadnaught’s size and firepower in the Nassau Class battleship with the Nassau joining the fleet in 1910.
The effect of FiOS on the state of long term expectations
In March 2007, 100 years from when Dreadnaught was named flagship of the English Home Fleet, the market reaction to FiOS is quite interesting to observe. There are those who would prefer that Verizon stop spending the capital on FiOS and return more value to shareholders through dividends and stock buybacks. This is a reasonable request by shareholders, but those with a long term view should be pleased that the anti-FiOS people are not leading the company. If you are a member of the “American Love it or Leave it” group, then you must certainly applaud Verizon’s decision. Despite the flow of good news about broadband deployments, the United States is still stuck in the Web 1.0 dialup world and falling behind other developed nation-states.
Here is what FiOS is really going to do to the market. For the past two years, competitors have been in denial that Verizon would have the capacity to continue spending on the project and they had hoped it would take longer to deploy. A higher cost and longer deployment time would anger shareholders and work in the favor of Verizon competitors. Unfortunately for competitors, the FiOS project is moving forward and this will force the cable service providers in the Verizon footprint to undertake their own widening of the Kial Canal project in the form of (1) spectrum upgrades sooner and (2) a complete upgrade of the CMTS 2.0 infrastructure.
Full Disclosure: I do not own shares of any of the companies mentioned in this post.