Fact-Checking Internet Association’s Amicus Brief
Politico reports that the Internet Association has filed an amicus brief in support of the FCC in the cases where Tennessee and North Carolina have challenged the commission’s attempt to pre-empt state laws restricting the ability of cities and towns to build broadband networks. The brief is light on the legal side, simply arguing that Section 706 of the Communications Act “provides that the Commission has the authority to take steps to accelerate broadband deployment that is not reasonable and timely.” The nature of the allowable steps isn’t addressed, but that’s of less interest than the claim that the nation is not building out broadband in a reasonable and timely manner. The claim strikes me as dubious.
IA’s argument depends on what the meaning of “broadband” is. The data from the National Broadband Map suggests that broadband is as close to universal as it ever will be if we define it as 10 Mbps down, with 96% of rural America covered and 99.9% of urban America. Tennessee is above average, with coverage to 97.3% of rural residents and 100% of urban ones. North Carolina has essentially the same coverage, with 97.2% rural coverage and 100% for urban dwellers. So there’s no argument to be made that Tennessee and North Carolina are failing to experience reasonable and timely broadband expansion at the 10 Mbps level. This figure is compelling because the FCC’s own “Measuring Broadband America” report in 2014 says 10 Mbps is the most meaningful threshold for broadband speed:
In specific tests designed to mimic basic web browsing—accessing a series of web pages, but not streaming video or using video chat sites or applications—the total time needed to load a page decreased with higher speeds. However, the performance increase diminishes beyond about 10 Mbps, as latency and other factors begin to dominate. For these high speed tiers, consumers are unlikely to experience much if any improvement in basic web browsing from increased speed–i.e., moving from a 10 Mbps broadband offering to a 25 Mbps offering.
But the Commission disregarded the assessment of its engineering staff and arbitrarily redefined “broadband” at the 25 Mbps level, based on the argument that common Internet activities actually require the speeds its engineering staff says are unlikely to cause a perceptible improvement in web browsing. Of course, it’s possible that other Internet activities can benefit from higher speeds, but if that’s the case we need to know what those activities are and how the improvement would be made perceptible.
Before we take up that question, it’s important to see what the impact of 25 Mbps vs. 10 Mbps has on the “reasonable and timely” question. While broadband is effectively universal at the 10 Mbps level (without even counting mobile broadband,) at the 25 Mbps level there are some gaps in coverage, especially in rural areas: nationwide we have 94% coverage in urban areas and 54.6% in rural ones. This divide reflects the fact that cable modem is essentially universal in urban areas but spotty in rural America: only 54.1% of rural America has a cable modem option, compared to 96.7% in cities and towns.
The scarcity of cable modem coverage in the boondocks doesn’t doom them to poor broadband, of course; America’s rural cable modem coverage is better than cable modem coverage in the entire United Kingdom, which stood at 48% for DOCSIS 3 in 2012 and only five percent in rural areas. Rural America relies on DSL, satellite, mobile, and Wi-Fi, all of which are capable of providing speeds of 20 Mbps or more.
IA’s argument depends on our willingness to accept the FCC’s definition because it fails to offer examples of applications that won’t work at 10 Mbps but will work at 25. The examples it does cite don’t make the argument that 25 Mbps is essential:
The services that Americans use are increasingly media-rich, requiring significantly greater bandwidth than in the past. These technologies include the deployment of 4K video, advanced gaming systems, and high-definition video conferencing services. Streaming video and audio, for example, accounts for 67% of downstream bytes during peak periods. And a recent report projected that by 2018, “digital TV and online video will be the two most highly penetrated [online] services, 86 percent and 78 percent respectively.”
4K video is a bust, at least at this point in its development, because of a dearth of content and a standards failure. To use 4K, you need not only a 15 Mbps Internet connection, you need a new TV and some 4K content, but the sales of 4K TVs are comparable to those of 3D TVs, the last “next big thing” that failed to capture anyone’s attention: disappointing. Advanced gaming is more latency-sensitive than bandwidth intensive, so it doesn’t fill the bill either, and high-def video conferencing is many times more important in the office than in the home. Streaming audio and video are nice, but they work perfectly well on a solid 10 Mbps connection.
IA also touts the importance of business broadband, but there’s no evidence that businesses are unable to get the kind of Internet connections they need. Business web sites are not located in homes or even in offices, they’re housed in hosting centers far away from business premises and these centers are awash with bandwidth. So IA tries the multi-user scenario to make its case:
The market for Internet-connected devices also has exploded over the last few years, and consumers with broadband Internet access are increasingly using multiple devices to access different services at the same time—e.g., streaming video onto a television while surfing the Web or checking email.
The multi-user argument depends on several video streams running at the same time: the typical Netflix rate is 4 Mbps, so a 10 Mbps connection can only support two of them at the same time that web surfing and email are going on.
So the “reasonable and timely” question comes down to whether we believe that the ability of rural Americans to view three video streams at the same time is essential for America’s prosperity. If we believe it is, it would also be necessary to determine whether the FCC’s steps are actually necessary given other dynamics that are currently afoot.
One such development is T-Mobile’s video streaming undertaking. ISP critic Brian Fung says T-Mobile is basically a cable company now:
T-Mobile is known for trying to shake up the telecom industry with its aggressive business tactics. Its latest, known as Binge On, lets you watch as many as 24 streaming video services without ever hitting your monthly data cap.
The channel lineup is pretty comprehensive: You can choose from HBO, ESPN, Showtime, Starz and Movieplex, not to mention Netflix and Hulu. Others could be supported in the future. You still have to subscribe to each of these services separately if you want them all. But the collection is big enough to make Binge On look like a compelling new way to watch TV. And this is the whole point.
T-Mobile is trying to become cable faster than cable can become T-Mobile.
Rural America has substantial terrestrial wireless coverage, 98.1%. If we assume the other mobile carriers will follow T-Mobile’s lead, as they did in unbundling phones from service subscriptions, it stands to reason that streaming over a combination of wired and wireless services will enable households with multiple teenaged kids to stream four or more programs at a time in short order.
There are things the FCC can do to help improve the state of rural broadband without destroying state sovereignty: it can accelerate the repurposing of government spectrum to cellular use, conduct a fair and reasonable spectrum incentive auction, and it can spend universal service fees more wisely.
If this subject comes up in at the FCC oversight hearing today, it would be interesting to learn why the FCC took such a rash step in terms of state and federal law when it has so many other options on the table. I’d like to hear an honest answer to that question, as well as those the AEI Scholars have posed and the one I posed in re/code yesterday. That would make for a fun hearing.