Biz & IT —

DC think tank tells Americans that their broadband is really great

Critics fire back: "But this isn't summer camp. This is our digital future."

Despite ITIF's conclusions, major cable providers like Comcast still dominate the American broadband landscape.
Despite ITIF's conclusions, major cable providers like Comcast still dominate the American broadband landscape.

Despite the fact that Americans are paying more per megabit than their counterparts in many European and Asian cities—a new report published by a Washington DC-based think tank says that broadband policy in America is totally acceptable.

The 76-page report (PDF) released Tuesday says that people who disagree with these findings “are holders of a particular ideology or economic doctrine, which is Neo-Keynesian, populist economic thinking in this instance.” Sure, the authors of this report—researchers at the Information Technology and Innovation Foundation (ITIF)—say that there’s “room for improvement in selected areas,” but they affirm that the “state of American broadband is good and getting better.”

Why would they come to that conclusion? After all, the ITIF’s conclusions are in contrast to previous findings published not just on Ars, but also by tech policy luminaries including, as the ITIF's report admits, “Susan Crawford, David Cay Johnston, Nicholas Economides, and Tim Wu and by think tanks and advocacy groups such as Free Press, the Open Technology Institute at New America Foundation, Harvard’s Berkman Center, and Public Knowledge.”

Those individuals and groups definitely have something to say in response.

“The ITIF report turns our national broadband policy into a self-esteem exercise, on par with one of those contests where everyone wins an award,” Harold Feld, of Public Knowledge, said in an e-mail to Ars. “'Hooray! We tried real hard and we're not so bad after all.' But this isn't summer camp. This is our digital future. If we want a world-class broadband infrastructure, we need to stop coming up with explanations for why things aren't really so bad after all and start dealing with the real problems right in front of our eyes.”

Intermodal, not just for transportation anymore

In its report, the ITIF outlines 10 major points as to why US broadband is awesome. We’ll take a few minutes to assess each point.

1. America enjoys robust intermodal competition between cable and DSL fiber-based facilities, with the third highest rate of wired intermodal competition in the OECD (behind Belgium and Netherlands).

First, “intermodal competition” is just a fancy way of saying that Americans have the option of buying Internet from their cable company or a traditional telco providing DSL service. Of course other countries don’t have intermodal competition—generally, European telcos came into being as incumbent monopolies were required to sell off their service at a given wholesale price.

“[As a result, European ISPs] could compete on price, but this competition was limited by the rate the phone company charged them for access,” Daniel Lyons, a professor at Boston College, told Ars. “And they could compete on customer service, billing, and other back-office processes, but these forms of competition don’t improve broadband performance and speeds.”

The US has a high level of intermodal competition—but other parts of the world have high intramodal competition. In practice, that means many Americans living in urban areas typically have a choice between Big Company A and Big Company B. We have a name for such a system: duopoly.

Case in point: I live in Oakland, California. I have a choice between AT&T’s U-Verse or Comcast’s Xfinity. Right now, I’m in the middle of Comcast’s “promo price” of $45 per month for 12 Mbps of service, although my real-world speed tests often show twice that level. Comcast’s DOCSIS 3 service is mostly always going to be faster—so in reality, it’s not much of a choice.

Sure, I could get satellite connection with data caps, for anywhere from $40 to $100 a month. If I lived in a different part of Oakland, I could even use the much-beloved indie ISP, Sonic.net. But the cold, hard truth is that in a major metropolitan area, I have exactly two choices between corporate giants.

At the end of the day, most Americans don't care whether they have a cable or a DSL provider. They just want good speed at a reasonable price, and they want a human being to actually pick up the phone when they call for technical support. Being intermodal is pretty immaterial.

Of course, not all telecom analysts agree here.

“Before concluding that a two-player market is competitive, we’d want to find some evidence of competition,” Lyons added. “Here, Comcast just finished an expensive rollout of DOCSIS 3.0 that gives its network a performance advantage over DSL. Presumably, if Comcast and AT&T were cooperating rather than competing, Comcast wouldn’t have paid for that upgrade. In response, AT&T announced a $23 billion push to deploy more fiber in its network—again, to improve its product and attract more customers. And the churn rates show that customers do in fact switch back and forth looking for the better deal—and no-contract plans make it easier for them to do so.”

While both companies may be competing with each other to some degree, there seems to be very little, if any, competition from anywhere else.

LTE won't replace your home connection anytime soon

2. America leads the world in the adoption of 4G/LTE mobile broadband, a technology that’s a credible competitor at the lower end of the broadband speed spectrum and a gateway technology for bringing broadband non-adopters online.

It may be true that LTE is slated to upend mobile data, but it’s still not there yet. LTE is still 2-3 times more expensive than a traditional home broadband connection. I've yet to meet anyone who has abandoned their broadband in favor of LTE service. Most mobile data services have significantly lower data caps than a home connection—and as a result, mobile data is not an adequate substitute.

While Ars’ Eric Bangeman had a good experience with a Verizon LTE mobile hotspot last year, he concluded, “This is the first wireless broadband device I’ve seen that could legitimately replace a wired home broadband connection, but doing so would be prohibitively expensive.”

3. Entry-level pricing for American broadband is the second lowest in the OECD, behind Israel.

So for the low-end of broadband, the United States has the lowest price? That sounds good, right?

But upon closer examination of the OECD’s most-recent data for December 2011 (XLS), low-end Internet users (defined as using the Internet no more than 10 hours per month, or 2GB of data) in the US have access to 3Mbps down and 512Kbps up speed—at $16.45 per month (purchasing power parity). By comparison, Internet users in Iceland have twice the level of service for $0.26 more. For less money, at $15.65, Hungarians get 10Mbps down and 1Mbps up.

We’re not really sure how the ITIF got this figure, and even if it is true, why it’s a good thing.

4. The average network rate of all broadband connections in the United States was 29.6 Mbps in the third quarter of 2012; in the same period, we ranked seventh in the world and sixth in the OECD in the percentage of users with performance faster than 10 Mbps.

ITIF, unfortunately doesn't cite its sources for the 29.6Mbps datapoint—it seems a bit high to us. Akamai, by comparison, reports that Q3 2012 average broadband speed in the United States (PDF) is at 7.2Mbps.

UPDATE 4:15pm CT: On a second look, the ITIF does cite the same Akamai report, although not on its first reference on Page 5 of the ITIF report, which is where we were looking. Later, on Page 41 of ITIF's report, the group clarifies that it's referring, in fact, to an "average peak broadband connection speed of 29.6Mbps," and cites that figure, accurately, to Akamai's Q3 2012 report.

Ars contacted Chris Nicholson, a spokesperson at Akamai, who agreed with us that the "average connection speed," of 7.2Mbps represents the actual, real-world average. As any Internet user can attest, real-world speeds fluctuate quite a bit, and an "average peak connection speed" is hardly the same as "average connection speed."

The ITIF argues, in footnote 88, that "In contrast to the average connection speed, the average peak connection speed is more representative of Internet connection capacity." There is obviously a great difference between real-world Internet connection speed, and Internet connection capacity. While the US may, in fact, have a high capacity, that capacity is clearly not being translated into actual products that a substantial portion of Americans are using. Further, it seems disingenuous to claim that the "average network rate" (which sounds a lot like "average connection speed") is the same as "average peak connection."

5. Of the nations that lead the United States in any of these four metrics (deployment, adoption, speed and price), no nation leads in more than two.

We don’t really know how to interpret this statement, nor what exactly it's supposed to communicate.

Do any of your non-techie friends have fiber? Exactly.

6. In the last few years American firms bought more fiber optic cable than all of Europe combined. 2011 was the first year in which America’s fiber purchases exceeded those of 2000, and 2012 orders have remained strong.

While fiber purchasing may be on the rise, that doesn’t necessarily mean that retail, fiber-to-the-premises is even available, much less priced competitively in most of America.

As Ars reported last June, despite test markets, lots of hype, economic stimulus, and lots of hope from bandwidth-hungry Americans, gigabit FTTP still hasn’t quite caught on across the country and is only even available in certain areas. Even wealthy Palo Alto, the home of Stanford University, couldn’t make it work.

And the City of Alameda, 32 miles northeast of Palo Alto, tried to run its independent fiber network for 10 years before selling the network, after notable losses, to Comcast in 2008. As of March 2011, industry estimates found that only about 7 million homes (out of about 125 million total American homes) have fiber connections, and most of those are from Verizon’s FiOS.

Sure, Google Fiber may be waiting in the wings—but for now, the service is only limited to Kansas City, Kansas and Missouri.

“Fiber doesn’t mean to the home—we have a big country,” Ben Lennett, the policy director at the New America Foundation, told Ars.

7. 82 percent of American homes are passed by a cable technology capable of supporting broadband speeds of 100 Mbps or higher and a new technology known as Vectored DSL may soon bring a second 100 Mbps service into the market.

Does that mean 82 percent of Americans actually have 100Mbps? Certainly not. As Ars has reported previously, 100Mbps is generally cost-prohibitive for most Americans—we generally pay far more money for far less service than our counterparts in other parts of the world.

“That same service in Hong Kong would be $37, and in Latvia [it] would be $31. In Seoul it would be $31 and Bucharest it would be $28,” Lennett added.

What problem?

8. Broadband adoption in the United States is not as high as some leading nations, but our 68.2 percent adoption rate for all households exceeds the EU-15’s 66.9 percent. When looking at adoption rate for households with computers, the US rate is close to the top (four percentage points from the leader) and three percentage points above the EU-15’s 85.9 percent for this population.

Again, let’s look directly at the OECD’s own figures (XLS) for “households with broadband access, 2010 or latest available year, by percentage.” The United States (at 68.2 percent) places 14th, behind Switzerland, Canada, Denmark, Norway, and South Korea at number one, with 97.5 percent. (We also find it odd that the ITIF would quote the “EU-15,” when the EU has 27 member states. The OECD does break out the entire EU as “EU27”—when Croatia accedes to the bloc later this year, it will become the 28th.)

It’s difficult to corroborate the ITIF’s claim of “adoption rate for households with computers”—which again, seems like an odd datapoint. It's unlikely that a household would have broadband access without a computer, unless an only-mobile data household would be counted. So does that mean Americans have a higher percentage of households with data-enabled smartphones?

9. American broadband service providers are no more profitable than those in the rest of the world.

While the profit margins measured for a telco as a whole may be relatively small (Comcast’s was only 7 percent in 2011), the margins on the Internet services portions of the company are “comically profitable,” according to one Wall Street analyst, Craig Moffett (though these figures don't necessarily account for the full cost of the infrastructure upgrades made by ISPs). Many American telcos, particularly the larger ones like Comcast, have lots of other business models besides simply providing Internet access. Comcast, for example, owns NBCUniversal, and its bottom line is included there, too. Foreign telecommunications companies like the UK-based Vodafone—which has a 45 percent stake in Verizon—stick primarily to traditional services, like Internet and mobile offerings.

10. American broadband prices are progressive: American users of low-speed, entry-level broadband services pay less than their peers in other countries, but those who use the fastest services pay more.

We know that this simply isn’t true—the OECD numbers bear this out. Furthermore, many broadband watchers continue to lament the level of mediocrity in America’s service.

“We have an endless series of reports over the last 7-8 years explaining why we should have low expectations and the experience of other nations should not tell us what is really possible,” added Harold Feld, of Public Knowledge, in his e-mail to Ars.

“None of this answers the real question: do we want to solve our national broadband problem or not? In the last few weeks, we have seen newspaper articles about children doing their homework in McDonald's because they can't get affordable Internet access in their homes. We have seen Fairfax County, Virgina, one of the wealthiest counties in the US, cancel its online textbook program because too many students had no broadband access in their home and could not access the textbooks online. Do we really want our national policy for these kids to be ‘sucks to be you, hope that McDonald's never drops its free WiFi?’”

Update: ITIF has posted a response on its blog.

Channel Ars Technica