Innovation Files has moved! For ITIF's quick takes, quips, and commentary on the latest in tech policy, go to

Wheeler Sets the Broadband Bar Higher than South Korea

Recently, FCC Chairman Tom Wheeler gave a speech arguing that “A 25 Mbps connection is fast becoming ‘table stakes’ in 21st century communications,” with the implication that anything less than 25 mbs is not really broadband.

This is an odd sort of statement, as it appears to be based not on any real analysis, but simply on the Chairman’s opinion. He tried to provide some rationale for this number when he stated “It’s not uncommon for a U.S. Internet connected household to have six or more connected devices – including televisions, desktops, laptops, tablets, and smartphones. When these devices are used at the same time, as they often are in the evenings, it’s not hard to overwhelm 10 Mbps of bandwidth.” I don’t know about you, but I personally am generally not using two devices at once. And as the Census Bureau reports, the average household size in the U.S. is 2.58 people with the median size being less. So, the majority of households are not overwhelming 10 Mbps of bandwidth.

So, if sub-25 Internet connections are not really broadband what does this mean in terms of what nations have broadband? By the Wheeler standard, it appears that not a single nation on earth has broadband. According to Akamai’s latest State of Internet report, the average broadband speed in Korea (the nation with the fastest broadband) is just 25.3 Mbps, but this is average so it suggests that it’s the median is likely under Wheelers’ vaunted 25 Mbps. Indeed, only 66 percent of broadband connections in Korea are faster than 15 mbps, with 37 and 33 percent in Hong Kong and Japan respectively. And you have to remember (as ITIF showed in our report The Whole Picture: Where America’s Broadband Networks Really Stand), these are three nations/regions that have among the highest urban population densities in the world and have provided among the largest subsidies to deploy high speed broadband. So if they don’t have real broadband the fate of humanity is in dire straights.

Akamai shows that U.S. average broadband speeds are 11.5 Mbps, faster than Canada and any other nation in the Americas; and 12th fastest in the world, and just 30 percent slower than Hong Kong the nation with the second fastest average broadband speeds in the world. Moreover, U.S. broadband speeds increased 21 percent over the last year, just 16 percent slower increase than the global average.

So what’s really behind Wheeler’s 25 Mbps line in the sand. In short, it appears that he is trying to move the goal posts around what is really broadband so he can make the argument that there is no competition for broadband and then justify government regulation, government-owned networks and other intervention to liberate our “Captive Audiences.” As he states “At 25 Mbps, there is simply no competitive choice for most Americans. Stop and let that sink in…three-quarters of American homes have no competitive choice for the essential infrastructure for 21st century economics and democracy. Included in that is almost 20 percent who have no service at all!”

Okay, I stopped and let that sink in. And after it sank in I said “so what.” First of all its wrong, because if you are measuring competition you have to measure the share of homes with only one provider, not one or none. And using this standard, the share is not three-quarters it’s two thirds. Second, this data is at least a year older than Akamai’s data and they show that U.S. speeds actually increased 21 percent in 2014. So presumably if the FCC used up to date data the share of households with only one provider of 25 Mbps would be even lower.

But more to the point, if Wheeler is going to argue that everyone needs at least 25 Mbps (at least he didn’t claim we all need a gig), it would be good to have some evidence. One place to look is to what people are actually buying when they have choices of speeds. For example, Verizon FIOS sells their 50 Mbps symmetrical service for 10$ more a month than their 25 Mbps symmetrical service. If our democracy depends on super-fast broadband, then presumably the majority of Verizon FIOS customers would be shelling out a mere 10$ a month more for double the speeds. While Verizon doesn’t release this data I would doubt that more than a small share of Verizon customers think $10 is worth it. I know I don’t. (We subscribe to the cheaper 25 Mbps service). Perhaps the FCC could ask Verizon for this data?

We also know that the share buying super-fast (and expensive) gig service is even smaller. As ITIF showed in its report, “The Whole Picture,” actual adoption of the gigabit broadband service provided by the municipal electric utility in Chattanooga, Tennessee (funded by federal stimulus dollars and priced at $300 per month), was less than a dozen subscribers as of August, 2012.

At the end of the day the Chairman is calling for more competition. Perhaps he should read ITIF’s “The Whole Picture,” which shows that the United States is third in the OECD in terms of the most intermodal broadband competition, with concurrent facilities reaching 85 percent of the population; only the Netherlands and Belgium are ahead, at 92 and 88 percent respectively. What more does he want?

Well clearly, he wants even more intermodal competition. Perhaps this is why federal broadband programs seems so committed to funding over-builders, rather than greenfield builds in places that don’t have any broadband. Perhaps he thinks that the prices Americans pay for broadband would actually be lower if through Wheeler policy providers (probably muni) were to build one additional wireline connection to all 130 million households in the nation. But if one more wire to every home is good, surely two more would be better, and three even better. So who would actually pay for the upwards of $175 billion dollars for each additional wire. Who would pay to offset the revenue losses of existing providers, who would on average lost one-third of their customers and revenue (but their actual costs would go down by perhaps 10 percent because the lion’s share of the cost structure of broadband providers is in the physical plant)? In the Wheeler universe, clearly consumers would pay none of these costs (although taxpayers may pay a lot of them). At the end of the day this kind of “competition uber alles” ideology is simply out of touch with the economics of broadband networks.

Print Friendly, PDF & Email

  • reader

    Many of these links direct to Robert’s e-mail login.