Fiber Economics

With the release of the FCC’s National Broadband Plan, Google’s announced intention to build gigabit fiber-to-the-home networks, and Verizon’s indications that they are not likely to be expanding their FIOS service to new areas, it’s a good time to review where we really stand with fiber.

The Real Reasons You Don’t Have Fiber

What are the real economics of broadband infrastructure? It’s not so simple as market opportunity, investment, and subscribers; Verizon and Comcast have different regulatory histories and see the world differently Google, as a potential new entrant, has completely different motivations.

Let’s take a look at the regulatory background, and then get a sense of what’s really motivating Verizon, Comcast, and Google.

Regulatory Background

We have gradually come to think of Verizon and Comcast as equals: big, for-profit telecom companies — competitors for TV, Internet, and telephone service. But they got to their current positions through very different routes. Here’s a brief (and rather incomplete) history.

In 1984, the former AT&T was busted up into seven Baby Bells: Ameritech, Bell Atlantic, BellSouth, NYNEX, Pacific Telesis, Southwestern Bell, and US West. Terminator-like, these companies have been spending the last 26 years reconstituting themselves, merging into very large firms. Bell Atlantic changed its name to Verizon in 2000 after acquiring GTE.

Telecommunications regulation in the United States has a long history and reflects theory originally applied to railroads and other public utilities. The idea was that communications was a public good and because the network had to be large and interoperable to be effective, it was best served by a natural monopoly. So, assets like public rights-of-way were made available for the monopoly to use, in exchange for an agreement to provide Universal Service, covering the entire population.

To keep the monopoly from charging unreasonable prices, regulators mandated that their services should be marketed at cost, plus a reasonable and sustaining profit margin. This means that there is no incentive for them to keep costs down; in fact, the higher their costs, the more raw dollars they make.

Verizon today operates under this kind of regulatory background, which was outlined initially in the Communications Act of 1934, and then amended by the Telecommunications Act of 1996 — which has subsequently been eroded and modified by case law and other FCC actions.

The FCC, under Bush-appointed Chairmen Michael Powell and Kevin Martin, tended towards the opinion that the best way to foster competition and innovation would be to empower a small number of well-capitalized firms and let them compete together in the marketplace.

Comcast, for its part, came together very differently. Cable TV franchises were primarily granted by local municipalities, starting in the 1950’s. Comcast acquired dozens of these small firms, each with their own regulatory agreements with cities and counties. By 2000 or so, this aggregation started to resemble the sort of “large firm” that the FCC thought could be an effective competitor to the telephone companies.

So that’s how we got here. Verizon is heir to the top-down, cost-based monopoly regulation subsidized by the Universal Service Fund, which requires that it provide telephone service even in rural areas. Comcast is the product of the roll-up of dozens (if not hundreds) of small cable TV firms. Now let’s take a look at their interests in the current landscape.


Verizon, in many ways, is just the current-day incarnation of a big chunk of the original AT&T. It’s still the primary telephone infrastructure provider and the bulk of its physical wiring plant is copper. It operates the same switching facilities that AT&T did back in 1984. In many important ways, nothing has changed.

What about FIOS? Isn’t Verizon innovating there? Aren’t they making this investment to “make money?” It’s complicated.

Verizon made the decision to install FIOS primarily to block competition. The Telecom Act of 1996 required that telcos make copper wireline infrastructure for competitors to run alternative services. This is where alternative telcos like Cavalier, Covad, Adelphia, and many others came into the market. You’ll notice that almost all of those companies are now defunct or severely hamstrung.

This is in part because Verizon (and its peers) set out on a strategy to make a competitive business model all but impossible. FIOS was part of that strategy.

When Verizon installs FIOS, they almost always remove the copper wires that could otherwise have been used by competitors; and this has effectively shut them out.

Verizon has spent over $20 Billion to build out FIOS in its service area. Ostensibly this might look like “investment in innovation” to observers. But in fact, this spending was mostly done to block competitors and to destroy the pro-competition provisions of the Telecom Act of ’96.

It should thus come as no surprise that Verizon has recently announced that they are unlikely to expand their FIOS network further. This isn’t because they can’t get more new subscribers in new areas (like downtown Boston, which is still not served); rather, it’s because they have calculated that the costs of future expansion exceed the downside risk of lost profits caused by competitors in the areas that remain.

They have put FIOS in all the places where it was either easy to do so or where the competition was too strong to ignore. Now that the competitors are mostly defunct or severely weakened, the threat is just not there to justify expansion.

Like feudal warlords, they invested just enough in FIOS to block out competitors, rejigger the regulations, and maintain a status quo of mediocrity. And we’re supposed to think this is innovation?


Comcast has different problems. Because all of their regulatory agreements are negotiated with individual municipalities, it’s more difficult for them to make investments across their entire footprint. This is why Comcast often rolls out new products and services in trial communities, and then rolls them out to new areas one at a time.

Comcast does have a very large television service footprint, and their acquisition of NBC and other content providers over the years, like HTS, is an attempt to establish themselves as a vertically integrated entertainment provider. They control the entire stack, from the physical cable, to the viewer, to the content itself. This means that they are protected from threats from content providers who might try to command high rates for popular content. Disney (who controls ESPN and ABC) often finds itself in battles about rates with cable providers. Acquiring NBC/Universal means one less potential threat of rate hikes for Comcast, and higher overall profits.

But Comcast’s physical plant is dominated by aging coaxial cable infrastructure. While local head-ends are fed by fiberoptic backbones, local distribution to the home is through co-axial cable, which can degrade in performance when it rains, is subject to lightning damage, and can only go so fast. Fiber-to-the-home is vastly superior, but it would cost Comcast billions to upgrade its plant. In the absence of competitive pressure (such as that which was faced by Verizon), they have no incentive to do so. Instead they are happy to push their existing plant as hard as it can go, using standards such as DOCSIS 3, and invest in fiber-to-the-home infrastructure only as necessary or convenient.


Google has recently announced that they would like to spawn innovation, and potentially build out gigabit fiberoptic infrastructure in one or more communities in the United States. I helped organize Baltimore’s municipal response to Google’s Request For Information for this project.

Google’s proposing something very different from what Verizon and Comcast offer: an open-access network, over which new entrants could provide Internet or other services. This is exactly the paradigm that Verizon has fought to destroy with FIOS.

Comcast has also fought open access repeatedly; before Verizon settled on FIOS as its primary anti-competitive strategy, Verizon tried to force cable companies to become subject to the same kind of infrastructure-sharing to which it was subject under the Telecom Act of 1996.

And Comcast fought this effort mightily; in 2002, I testified before the Maryland House of Delegates in support of a bill that would force Comcast to open its network, and Comcast’s lobbyists managed to defeat it.

Also in 2002, working alongside Verizon-supplied lobbyists, I testified before the FCC with TCP/IP co-inventor Vint Cerf (now a VP at Google) arguing that cable companies should be forced to provide “open access” to their networks because it would promote competition and entrepreneurship. At that hearing, FCC Commissioner Robert Pepper made it clear that the FCC believed that Verizon and Comcast could provide all the competition we would ever need. We see how that’s turned out.

To date, there has not been any significant open access network deployment in the United States. And with the decline of competitive telco-based services, telecom innovation has now stalled entirely. It’s time for something new.

Net Neutrality

Google has a different potential problem on its radar. In the US, Comcast and Verizon control access to a large percentage of its customers. Currently, the Internet operates under a doctrine of “Net Neutrality,” which is to say that customers and Google all just pay for access to the network, and each can communicate freely with anyone else on the network.

Various telecom executives (most notably former AT&T CEO Ed Whitacre, now CEO of GM) have argued that companies like Google should no longer get free access to their customers. Folks like Whitacre believe that the natural role of a company like Verizon or Comcast is to act as a toll-gate, charging both content providers (Google) as well as customers for access to each other.

As you might imagine, Google heartily opposes this idea: it could dramatically increase their costs and would destroy the “level playing field” which has dominated the Internet from the beginning. Startups could be stifled because they might need to negotiate an agreement with broadband providers to get access to customers. This is a war that Verizon and Comcast appear ready to start, and people like News Corp’s Rupert Murdoch are fanning the flames.

Google’s Fiber Plan

Google’s announcement that it intends to build ultra-fast open access fiber networks is its declaration of war against the threatened end of net-neutrality. Further, this is a productive use of Google’s vast stockpile of cash; it’s something tangible it can do to ensure its market position.

And it’s a move that’s ideologically compatible with its mission. Google genuinely believes that the expansion of a fast, net-neutral Internet has positive effects on society, and it’s also good for its bottom-line. More people online means more ad-views which means more advertising, and more dollars for Google. There’s no downside for them; it’s an expensive proposition to be sure, but it’s less expensive than paying for access to customers in a world without net-neutrality.

By promoting itself as a good citizen, wrapped in the banner of open-access, innovation, and net-neutrality, ideologically-sympathetic regulators such as the FCC’s new Obama-appointed Chairman Julius Genachowski are likely to view Google’s approach favorably. This would allow Google to establish a vertically-integrated, long-term market position which would be hard for Verizon or Comcast to disrupt.

And the kicker? The open-access network Google’s proposing really would promote innovation and entrepreneurship. The United States is ranked 15th in broadband penetration worldwide today. This is a chance to change that.

Don’t believe that Verizon or Comcast will make these investments unless forced to do so. And while Google may also feel it has no choice but to build its own network, Google at least has a vision that goes far beyond just sustaining a mediocre status-quo; they truly believe in the level playing field that has given birth to so much innovation.

It’s time for America’s bandwidth to finally match our ambitions and our talent. Let’s go Google!


#1 bryanl on 03.27.10 at 7:25 pm

This is a really good piece. It is scary what Verizon and Comcast will do to try to stifle innovation. I'm happy to see you championing the Google FIOS initiative in Baltimore.

#2 Joe LeBlanc on 03.27.10 at 7:58 pm

Excellent post. One minor detail I'd like to get clarification on though: in what market(s) did Adelphia operate as an alternative telco? Growing up, we had Adelphia as the cable provider and they were the only ones providing cable service (to my knowledge). Did they also operate as an alternative after the 1996 Telecom Act, or were they always an alternative telco that just so happened to have exclusive rights in some markets?

#3 davetroy on 03.27.10 at 8:23 pm

Adelphia operated as an alternate voice and data telco in Baltimore, at least, until they went defunct with their fraud issues in the 2002-2003 timeframe. I know, I had to threaten to sue them to get them to correct $60,000 in over-billings.

#4 Mykel on 03.27.10 at 9:59 pm

In the intro, you talk about how Comcast isn't spending billions on rolling out an all-fiber network because a) they really can't on a large scale b) they don't have incentive to.

With Google providing an open fiber network, would that enable Comcast (and/or Verizon) to effectively “hijack” the fiber lines for their own business?

Meaning, could Comcast suddenly say, “Get Comcast Fiber in your area!” which is actually just a rebrand of Google Fiber lines?

#5 Dennis D. McDonald on 03.27.10 at 10:03 pm

I was wondering why I can't get FIOS!

#6 davetroy on 03.27.10 at 10:04 pm

In theory Comcast, Verizon, or any other company could offer services over Google's physical plant. It's hard to say what the terms for that might be as it's new regulatory terrain. But this is what Google's fighting for. For one thing, it would give Google control over Comcast, not the other way around.

#7 Rasul Sha'ir on 03.27.10 at 10:59 pm

Very interesting and informative read David. Great post. Thanks for the the fiber economics history lesson. Good stuff.

#8 ecogordo on 03.28.10 at 1:17 pm

Let's hope that the Feds understand how important Google's effort is towards driving innovation which in the end keeps us competitive globally.

While the US is currently the content king which keeps our head above water, it seems to me that without Google's stake in this fight, the status quo will not make it for us over the long term.

Another part of the equation is productive engagement. While we all want these investments to be made, perhaps we also need to consider how productive we can be with what we already have given the state of the economy.

When I look at, I am amazed at what we have. How do we leverage these strengths to build a better, stronger community?

#9 Max Efremov on 03.28.10 at 7:44 pm

That was a damn excellent post David. Two questions:

1) Isn't Google's move, entitled GoogleFiber, only a trial run? Similar to the deployment of WiMax wireless service in Baltimore, wouldn't GoogleFiber just be a one-city, one-service thing? What exactly are you doing when you “helped organize Baltimore’s municipal response” to Google? Is Baltimore trying to win some sort of raffle, albeit certainly less random than a name out of a hat?

And question 2) What happens if Baltimore wins? Or if loses? Will Google roll out their “FiOS” nationwide in the coming years? Or will they only do densely populated cities, counties, etc?

#10 davetroy on 03.28.10 at 7:55 pm

Thanks Max.

1. Google is conducting this as a trial, in order to learn what it would be involved in replicating it elsewhere. The only reason they would really want to try this is so they can scale it up elsewhere. So we would have to assume that's the plan, unless it goes very badly or circumstances change dramatically. If Verizon decides to react by building something highly competitive, Google might decide that it's not necessary for them to act.

2. Regarding Baltimore, there is only upside. If Google chooses to work with Baltimore, they will begin implementing their proposed network this year. If Google chooses not to work with Baltimore now, they may choose to do so later; also, we have highlighted in clear terms the need for this kind of investment here which other carriers may decide to act on. Alternatively, we could build a gigabit municipal fiber infrastructure ourselves — and then make it available to Google later, earning returns on it for the city.

One would presume they prefer the density only cities can provide. After that, it would be reasonable to assume they would consider deploying in contiguous areas such as the counties.

However, I could easily envision a universe where only dense urban cores are served — raising serious long-term questions about the viability of the suburbs. You heard it here first.

#11 Why is Google building a fiber network now? « Google Fiber Asheville : Official Website on 03.29.10 at 10:45 am

[…] A great post that explains the regulatroy landscape and the actions of individual broadband companies: On “Fiber Economics“ […]

#12 Why Google & Verizon Won’t Be BFFs Forever - A Collection of Latest Happening in Technology Field on 03.31.10 at 12:45 am

[…] business decision to get ahead of the competition when it comes to broadband speeds, while making sure it doesn’t have to share its pipes with others. Verizon has in fact slowed its fiber expansion, and notably, has a history of dumping the lines […]

#13 Fiber Economics — Dave Troy | Finshake on 03.31.10 at 5:27 pm

[…] Fiber Economics — Dave Troy: Fueled By Randomness. […]

#14 Localhost on 03.31.10 at 10:41 pm

NN section entirely out of date. As one example, Ed Whitacre now runs GM.

#15 Russ Nelson on 04.06.10 at 5:18 pm

Please don't call it Google FIOS. FIOS is a Verizon trademark for their crappy 20/20 fiber service (yes, 20/20 is relatively good, but technologically crappy).

#16 Russ Nelson on 04.06.10 at 5:22 pm

Yes, they could. So could you. And they'll know that. If they attempt to charge much more than commodity bandwidth prices, they'll have competition. That's the whole point behind open access — it eliminates the last mile as a competitive point.

#17 bryanl on 04.06.10 at 5:46 pm

Totally didn't mean to call it FIOS. I really do hope Google can do good things in this area. I have Verizon FIOS 50/50, and I've been impressed with their service. I also don't live in the city, so I'm sure the point is moot.