Blog Post

The Digital Divide Will Ensure a Broadband Ghetto

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

If you live in New York City or in any of the heavily populated and wealthy areas of the Northeast, you likely have access to some of the fastest broadband speeds available in the country. If you live in a suburb of Austin, Texas, however, you’re offered speeds some six times slower for about half the price. And as the technologies race ahead for network access, ISPs with fiber to the home and cable-provided Docsis 3.0 service are going to surpass the speeds that providers using old-school copper and even wireless can offer.

Which means that while some people will be living in the 21st century, great chunks of the country will be subsisting on the 2010 version of dial-up. Don’t believe me? Verizon has tested a tech that requires software upgrades to deliver a 10 Gbps connection that’s shared among 32 homes. Meanwhile, technologies such as the next generation from AT&T are going to require swapping out gear at the node and inside your home so the companies still using copper can squeak up to speeds that most analysts doubt will even reach 100 Mbps. That means a hundred-fold disparity in connection speeds. Holy cow.

We’ve addressed this in stories, when FCC chairman Julius Genchowski came to visit our offices and even in conversations with service providers. It’s the key reason the U.S. isn’t competitive with other countries when it comes to the rapid deployment of next-generation networking technology. So we think it’s a big issue. The FCC even thinks it’s a big issue. Google thinks it’s a big enough issue that’s it’s going to spend hundreds of millions of dollars to build a shared, experimental fiber-to-the-home network — and embarrass the ISPs and the government in the process.

To quote the Google blog: “After all, you shouldn’t have to jump into frozen lakes and shark tanks to get ultra high-speed broadband.” Oh, but that’s what people are willing to do because the ISPs are so focused on the short-term bottom line to spend billions (yes billions) on big pipes. When I talk to many ISPs, the No. 1 reason they give me for the lack of investment in faster infrastructure is that users don’t want it. There’s no killer app driving demand.

I’d buy that if ISPs were investing in research to find those killer apps, or weren’t actively seeking to whittle down connection speeds with tiered pricing plans or caps. But consumers are finding killer apps of a sort — only those killer apps interfere with the pay-TV side of the equation for cable guys and telcos that have invested in IPTV infrastructures. But even if the idea of another we-have-more-HD-channels-fight thrills you, the truth of the matter is that competition in broadband speeds will bring about the connected home.

No longer will just your computer always be on; your thermostat, your television, your appliances and your links to loved ones will be, too — and possibly throughout every room of your home. But that’s not going to happen for those without fat pipes. Which is why I and others found the National Broadband plan so disappointing. While it gets a lot of things right when it comes to boosting broadband adoption, it doesn’t really help drive innovation or ensure competition.

And the plan’s authors are aware of this. They offer the salve of data as a means to possibly protect consumers from overcharging and to shame providers into offering better service, but that’s going to require an activist FCC like the current commissioners are, and in a few years, who knows if that will be the case. And I am as excited about faster mobile broadband as anyone (maybe more), but mobile data isn’t going to replace wireline broadband — nor should we be content with those speeds.

I understand that having spent $19 billion building out a fiber network, Verizon would be justifiably upset over sharing its fiber with competitors. We also currently have a huge variety in our network providers that makes sharing lines difficult from both a technical and regulatory perspective. However, the type of open-shared network that Google is proposing is the fastest ways to get the U.S. big broadband quickly.

Since the FCC can’t get it done through the Broadband Plan (it isn’t even trying) and Congress decided not to get it done with the stimulus money (a few fiber projects were approved), then it’s up to the FCC and Congress to make it easier for cities to deploy their own fiber networks which can then be opened up to competition.  That means making it possible to change the laws in the 18 states that make it difficult for municipalities to own fiber networks or offer competitive communications services, but will likely evolve as a states rights battle.

Right now, Google and consumer demand is our best hope for better broadband, which isn’t making me feel too great about the future. Anyone have a freezing lake I can dive into?

Related GigaOM Pro Content: Who Will Profit From Broadband Innovation? (sub. req’d)

Image courtesy of Flickr user Tycho Moon

29 Responses to “The Digital Divide Will Ensure a Broadband Ghetto”

  1. As usual, I have questions about who to believe. I haven’t studied this much, and it seems no matter what the subject, there are always many “with the facts” who actually take those facts out of context and reinterpret their meaning or slant them one way or the other for whatever purpose…

    So I don’t know whether Stacey is right or all her critics — but here is what I do know: I am in Texas. I have broadband. It costs enough. And I have yet to find a broadband service here that could be called remotely reliable and that can consistently live up to the given company’s sales hype.

    Some may need more speed, but what I’d really like is to find, at reasonable cost, the speed and reliability I’ve already been promised.

    Stacey, do I need to move to New York for that?

  2. Martha Hyde

    I find the pie chart above interesting with 78% having only 2 providers. I strongly suspect that for many homes with “2” providers, those 2 providers are in actuality only one. For instance, AT&T and Suddenlink provide my community with broadband. However AT&T owns Suddenlink and “both” companies offer exactly the same plans at exactly the same costs. If the FCC claims that 95% of Americans have access to broadband, that seems to be a bit skewed since by their own assessment this year that 50% of the rural areas do not use or do not have access to broadband. Access means you can afford it if available and most people in rural areas earn far less than those in the cities, meaning that broadband access, if available, is usually priced out of their means. Furthermore, many may not use the internet at all because dial-up is so incredibly much slower than it was when it first came out, precisely because of the demand for greater bandwidth by those living in cities. One internet TV watcher can displace 30,000 dial-up users because of the sluggish internet highway.

    For that reason I support ISPs charging more for greater bandwidth use. Our internet highway has reached the state of not being able to handle any more users as it is. However I do not support ISPs being able to limit access to particular websites, even if they are high bandwidth users, as long as the user pays for them.

  3. Stacey, if the comments here are an indication of the challenges that the FCC will encounter in its quest to attain superior infrastructure — or more accurately, an objective of broadband “parity” with the global market leaders, then progress in the near-term for U.S. major metro areas seems unlikely.

    That said, perhaps setting a lower bar of expectations is more realistic — say 10Mbps symmetrical broadband access in the Top 10 U.S. metro areas in 10 years. Meanwhile, developing nations will have the opportunity to catch up with the U.S. and attract more investment and create new high-tech jobs in their rapidly evolving economies.

    Perhaps this is what Mr. Walsh and others are suggesting in their commentary — America has an opportunity to be benevolent, by enabling other countries to reap similar economic benefits that have been bestowed upon the U.S. economy for decades.

    Truly, this will be a wonderful gift for developing nations. In particular, technology-centric business leaders in China and India will no doubt be thankful that their counterparts in the U.S. market have decided to settle for “good-enough” broadband.

    • The average download speed of the highest speed service tiers in use in America’s ten largest cities today is 25.67 Mbps, per measurements. I don’t think reducing this rate down to 10 Mbps would be in the nation’s best interests, David.

      New York 58.19
      Los Angeles 20.21
      Chicago 26.39
      Houston 25.72
      Philadelphia 18.55
      Phoenix 20.01
      San Antonio 12.59
      San Diego 30.66
      Dallas 21.89
      San Jose 22.49

      Go to and see for yourself.

      • Richard, understood, the selective use of market data produces some very interesting results.

        However, as I’m sure you know, the highest speed (highest price) service tiers represent a small fraction of the total broadband subscribers in each of those cities.

        In contrast, I’m suggesting raising the expectation for all “mainstream” subscribers, and with symmetrical data rates (10Mbps up/10Mbps down) — to gain some degree of parity with the global market leaders.

        Let’s not delude ourselves any longer, but instead agree on an acceptable level of mediocrity that all U.S. stakeholders will support as a meaningful short-term objective for progress.

        We can do this together, but first we must move beyond denial.

      • Mr Deans says: “…the selective use of market data produces some very interesting results.”

        The data I presented are measured speeds of actual users in actual markets. The data show the fastest services that are available and are being used. This is the real truth, in other words. This is simply by way of taking stock.

        BTW, nobody wants a symmetrical service, so it’s a waste of money to design networks to provide it.

  4. “the truth of the matter is that competition in broadband speeds will bring about the connected home.

    No longer will just your computer always be on; your thermostat, your television, your appliances and your links to loved ones will be, too — and possibly throughout every room of your home. But that’s not going to happen for those without fat pipes.”

    Using this well worn nonsense to try and justify more gov’t interference in internet access is pathetic. The internet enabled appliance is something no one wants except maybe an overreaching central government bureaucracy that wants the ability to turn off your appliances after “cap & trade” legislation destroys the US energy grid.

  5. This thread of comments is extremely frustrating as it’s missing the point, namely that for many Americans bandwidth capacity is inadequate and for all Americans bandwidth is too expensive.

    To Jack’s points, how can we claim that America’s broadband growth has been “mind-boggingly fast” when you can get so much more bandwidth for so much less money in other countries?

    And I don’t know where you get this statement from: “The ‘2010 version of dialup,’ as Stacey put it, is dozens of MB/s…” There are huge swathes of America that can’t get 10Mbps, let alone dozens. And that’s putting aside the factual point that “dozens of MB/s” would actually mean more than 100Mbps as 1MBps equals 8Mbps.

    To Kevin’s point, why are you comparing the cost of bandwidth today to what it was in the past? Isn’t it more important and relevant to compare the cost of bandwidth in America compared to the world’s broadband leaders today? What would have happened to America’s position as a global economic leader in the 20th century if electricity was more expensive and less available here than other countries?

    While I’m not necessarily a fan of government regulation and I do think it’s important we acknowledge the progress that industry made on its own without a national broadband plan to guide and support them, but to say that America has no problems when it comes to the cost of bandwidth suggests a lack of understanding about what really matters.

    To Brett’s points, just look at ITIF’s report called “The Need for Speed” to see a very plausible use case for how the average home of the near future is going to need at least 90Mbps symmetrical to use the applications we know about today, let alone the apps that may come online tomorrow.

    And to your question of “are you willing to PAY for all that expensive bandwidth?” Why aren’t you looking at this as an opportunity to lower the cost of that bandwidth? I know that’s not necessarily a simple thing to do as it requires a lot of investment in new fiber networks, but if we just accept the current paradigm as the way things are then you’re right that we’ll never be able to achieve a 100Mbps future as the high cost of backhaul will always be a barrier. But that gets to the heart of my point that we need to find ways to lower the cost of bandwidth, and that’s going to require significant investment in new capacity to increase the supply and lower the cost of bandwidth.

    And to Richard’s point, while you’re likely right that Stacey’s specific example may be limited by the economics of deployment, do you really feel comfortable as an American claiming that just because an area’s difficult to deploy to that it’s not worth the effort? Now I know we can’t ignore the economics of deployment issue as then we risk talking in feel-good terms rather than dealing reality. But I personally have a problem with the idea that deploying next generation networks is too hard and therefore we can’t or shouldn’t do it.

    If we think America needs a certain level of infrastructure to be globally competitive in the digital economy, then we should be moving heaven and earth to equip our communities with that infrastructure.

    I sometimes wonder what would have happened if our forefathers had had the same can’t-do attitude as many seem to have about broadband, what America would’ve looked like in the 20th century. What if 100 years ago we chose not to electrify the country because we weren’t sure people were going to want it, or we got scared by the cost and complexity of accomplishing this admittedly huge task?

    Finally, to simply say that market-driven competition is the panacea to all our broadband woes is dangerously false. For starters, it’s obvious that for most of rural America the economics will never work to enable robust facilities-based competition. But even for those communities with better demographics, the inescapable reality of broadband competition is that it’s concentrated in the most economically viable areas. It makes more sense for profit-driven corporations to focus their resources on serving the areas that have the best chance to generate a robust return on investment. Also, we’ve already seen that competition does drive investment but only in markets where it exists. For example, the majority of DOCSIS 3.0 deployments have occurred in cities that have fiber. What this means is that limited capital is being ever more concentrated in the richest markets. And because capital for deployment is limited, that’s inevitably creating the broadband ghettos that Stacey’s mentioning.

    Unfortunately, what this points to is that Stacey’s claim that competition is the key to driving higher capacity lower cost broadband service is also false. For a new facilities-based wireline entrant to achieve breakeven in a market they need to reach take rates of around 30%. If there are already two other wireline providers, which there are in most markets, then if the new entrant is able to achieve those take rates that most likely means customers were taken from the incumbents, lessening the capital they have to invest in upgrading their networks. And simple math shows that even the best markets will have trouble supporting even three facilities-based wireline broadband providers.

    To those that claim wireless will provide that competition, if the world is heading to 100Mbps speeds, then we need to face another reality that wireless can’t handle that, especially if we’re talking about delivering an actual 100Mbps rather than an advertised.

    My final point for now on these matters is that we need to stop focusing so heavily on “competition” when what really matters is the capacity, price, and performance of service. What’s wrong with having a broadband monopoly if that monopoly provider is delivering tons of cheap, reliable bandwidth and continuing to invest in upgrading their network?

    I’m not saying that monopolies are the answer or that competition is a bad thing, just pointing out that what really matters is that Americans have access to the service they need to take advantage of all that the Internet has to offer.

    • Interesting comment, Geoff.

      The more relevant ITIF report re: Stacy’s complaint is “Explaining International Broadband Leadership,” It separates the policy factors such as unbundling and subsidies from the non-policy factors such as urbanicity and demographics. The report finds that 75% of the variation among nations in broadband speed, penetration, and price is non-policy related. Paris has 2100 km of sewers that permitted that city to be wired for fiber at very low cost, so the residents reap benefits from a civic project that began in the 14th century, when North Americans were still shitting in holes in the ground. It’s hard to make up for a 600 year head start. The rest of France, BTW, is a DSL ghetto with lower average speeds than the US.

      Overall, the US has higher broadband speeds than Europe, and we only fail to compare favorably against the small pockets of fiber in a few large urban centers such as London, Paris, and Amsterdam. Sweden started a huge fiber program 10 years ago at taxpayer expense that’s impossible to duplicate in the US given our budget problems and higher-priority needs in health care and education.

      95% of America is already wired with a system that provides 4 Gbps to each home: it’s called cable TV. Americans prefer to use most of this bandwidth for TV programming, but it’s shifting to general-purpose Internet access as consumers demand it. The transition of this system from digital cable to Internet will happen in the fullness of time, and will do so at the rate that the average American wants it to shift. The cable companies aren’t run by stupid people, they’re run by people who chase the consumer dollar wherever it’s to be found. The transition isn’t happening as fast as most of us geeks want it to happen, but we’re a minority.

      Does this system need to be replaced by fiber? At some point, sure, but that point is very far off in the future, like 20 years away. For the immediate future, HFC can meet the needs of most people, supplemented by mobile broadband (because it’s mobile, not because it’s as fast as FTTH.) Deeper fiber penetration into HFC, FTTC, and mobile networks is happening as we speak, to the tune of some $60 billion/yr.

      Who’s in control of the rate of fiber rollout, the network operators, the government, or the people? I would maintain that the rollout is already far ahead of the services that most Americans are willing to purchase. The average American consumer opts for lower speed, lower cost Internet access over faster and more expensive options every time.

      America leads the world in low-cost, low-speed broadband, and this is one of the things holding us back. Every municipal broadband system that’s tried to go head-to-head with DSL and cable has racked up massive debt and failed to deliver on its promises because the demand isn’t there.

      America will have faster and cheaper broadband when consumers demand it, and right now they aren’t. So don’t point the finger at the operators or the FCC, look in the mirror.

  6. There are parts of the Austin suburbs that are very challenging to wire on account of their being satellites of development surrounded by undeveloped property. Essentially, there is a ring of subdivisions within 40 miles or so of downtown that are considered commuter territory because Texans aren’t averse to driving long distances to get to work.

    Given how personal you’re making your complaints about your broadband service, Stacy, why don’t you post a Google map of the subdivision where you live – home address is not necessary, just pick a generic address – just so people can get an idea of how representative of Austin your neighborhood is. Austin does have some U-Verse and Road Runner Turbo neighborhoods, so it would be useful for your readers to understand why these services are not yet available in your neighborhood.

    If you’re outside of the Austin city limits, as I think you are, an Austin muni network isn’t going to do much for you in any case. For people who live in the boondocks, the best immediate solution for higher speed broadband is to move to an area where there’s more willingness to pay for premium services.

  7. I have to agree with Kevin’s comments. The growth of Internet connectivity in the US has been mind-bogglingly fast, and it shows no signs of stopping. The “2010 version of dialup,” as Stacey put it, is dozens of MB/s — far more than anything that was available just a decade ago.

    Scott Wallsten has done some excellent research ( on the validity of international comparisons, and he points out serious flaws in the metrics that claim to show the US falling behind our peers. One of his most compelling points is that, no matter what you think about speeds and prices, Americans are getting much better use out of their Internet infrastructure in terms of music and video downloaded.

    But all that aside, I want to specifically object to Stacey’s conclusion. “Right now…consumer demand is our best hope for better broadband, which isn’t making me feel too great about the future.” Really? Internet connectivity is continuing to skyrocket under anyone’s projections. More importantly, consumer demand is precisely what’s driven all this growth so far. What in the world could have driven it, if the demand weren’t there? Moore’s law hasn’t kept up for so long because of any FCC plan; it’s kept up because Intel and AMD have customers who’ll pay for speed and who still want more. The law of supply and demand has been building our networks just like it’s been building our processors, and there’s no reason to expect it to quit.

  8. Judith Nappe

    The mess of broadband availability in rural areas in the Pacific Northwest is never going to get good, it seems. I have a fast computer and cannot afford the outrageous price for hi-speed broadband so have to settle for the lowest speed the monopoly phone company offers: 768 kbps. Very slow. They have copper wire out to my little town, and as a monopoly, have no competition because they keep local providers from reaching us. It is sickening when I remember there was better telecommunication service in the 1950’s. Century Link is not doing any upgrading and probably never will. What a backward country America is.

    Hint: when commenters are talking about speeds of broadband, you’d do yourself a favor to state the kilo bits/sec or gigabits or whatever you have is called. Gives readers a base to compare.

  9. it’s articles like this that make me question our grit as a country. stop complaining. if there is a REAL business case, prove it, get funding, and build the darn thing yourself.

  10. We don’t need the government to invest in commercial goods. They are bankrupt and make bad investment decisions.

    The fact that Google and Verizon are invest Billions in Broadband simply reinforces this notion in my mind.

    Cut government spending, and cut government involvement.
    Save the money for Social Security and Medicare.

  11. Hmmmm…
    Do you suppose a community willing to pay for and install fiber through-out its’ citizens abodes would induce a quid-pro-quo from local providers ? A stray thought only.

  12. I’ve followed GigaOM for years, but I can pick out Stacey’s pieces from a mile away these days. They will have an intently Texas slant (as far as I can remember, she’s the only writer who sees the need to push a local agenda), and they will probably be re-iterating for the thousandth time that broadband in the US is slower than other countries (a falsehood), and calling for the FCC to regulate it harder. The difference between blogs and published magazines is the editing. There’s no way an editor would allow a writer of a published newspaper or magazine to keep re-iterating the same topic over and over and over again.

    On the actual subject, businesses routinely have gigabit to every desktop with tens if not hundreds of megabits between locations, higher than the average consumer has. They have the funding to procure technologies out of reach of the average consumer, yet they have not found a need for larger bandwidth aside from video. Where’s the killer app? If there’s a demand for increased bandwidth, there’s enough competition that someone will do it. I don’t buy the oligopoly argument here, given the process we’ve made in broadband penetration and speeds in the last 10 years.

    There’s no reason to re-invent digital cable using IP, the distribution system we have in place works. There was a compelling reason to re-invent voice, as TDM networks were inefficient relative to packet networks for individual to individual communications, but broadcast networks are at least as efficient, if not more efficient, as IP (given that multicast is not implemented Internet-wide and is a confusing technology to implement even on an enterprise or metro network) at delivering video to thousands of subscribers. I’m OK with the FCC’s plan, I don’t get the calls for additional regulation on top of the proposals. This blog is starting to carry more opinion than fact these days.

    • A fine comment. If I may pile on, how can Texas broadband speeds be 6 times slower than NY? If they were one time slower they’d be stopped. Oh, yeah, dramatic license trumps accuracy.

  13. Brett Glass

    Stacey, before you whine that “Ooooooh, I can’t get the absolutely fastest possible broadband,” you need to answer two questions. Firstly, what do you want to do that really requires that much capacity — and isn’t simply wasteful? And, since bandwidth costs money (especially way out there in the middle of Texas), are you willing to PAY for all that expensive bandwidth? Put up or shut up: unless you can answer both of these questions, you have no right to bellyache.

  14. Ok, this is getting out of hand.

    According to the FCC’s opening bid to regulate broadband (otherwise known as the National Broadband Plan), 95 percent of Americans have access to broadband, and 85 percent of those people can chose from multiple suppliers. The FCC pegs broadband adoption at over two-thirds. Research firm figures this to be over 70 percent. According to Pew Internet, the average price of broadband in 2009 was $39, or one-fortieth as much as dial-up on a per-bit basis only a decade ago. Finally, the FCC estimates that the average actual downstream broadband speed is about 4Mbps which, according to Akamai, is a shade higher than the average for European countries, all of which are smaller and denser than the US and have more intrusive regulatory regimes.

    Do these facts really paint a picture of a “broadband ghetto”? A “digital divide”? Apparently the FCC thinks so and is quietly setting that stage to get its regulatory mitts around broadband operators (that’s essentially what the USF shift is all about) so that it can fix this massive problem.

    Keep in mind that the strides made in broadband deployment during the last decade (and for those that weren’t taking notes, broadband adoption was less than 5 percent in 2000) occurred precisely because broadband was largely unregulated. In fact most progress was made subsequent to the FCC’s 2003 Triennial Review which largely undid the failed experiment embodied in the 1997 Telecom Reform Act. All of this was done with private capital and with private innovation.

    Yes, it would be nice to have broadband available to 100 percent of Americans. And yes, we’d all love 100Mbps broadband connections. But we should be instinctively wary when government steps in to regulate or reform industries which heretofore have operated fairly well.

    • Kevin, broadband adoption increased by so much in such a short period of time because the mainstream found a use for faster speeds in services such as iTunes downloads and more graphically intensive web pages, but the investment in DSL and cable broadband was already made. So the speeds capable for delivering those services were there in many areas of the country. Today people are finding they need faster speeds for video streaming, HD downloads and video uploads, but they are running into issues with those things taking a lot of time of being of inferior quality. The speeds aren’t here today.

      • Stacey, fair points although it might also be argued that capacity and demand inch up in tandem with each other, much like computer memory and applications software. Demand drives capacity increases, but capacity increases also enable whole new sources of demand.

        Also, streaming applications will always encounter challenges in best-effort networks, almost irrespective of the speed of the last mile connection. When everyone has 100Mbps fiber to the home they’ll all be watching multiple HD streams and the aggregation (“middle mile”) network will still be heavily congested. Streaming applications operate best when the network can provide some level of QOS, something that many broadband critics strangely view as evil.

    • I assume you are a rep. and think that company’s should be allowed to run there company any way they want with no regulation what so ever, well if that’s the case then you can keep on paying more than double for half the speed.