An updated report by the New America Foundation (NAF) examines whether we are getting a good deal on our broadband in the United States. It does so using fairly straightforward methods: cataloguing advertised prices and speeds for major cities around the world. Unfortunately, to paraphrase H. L. Mencken, for every complex problem there is an answer that is clear, simple, and wrong.
We covered the issue exhaustively in our report from earlier this year, The Whole Picture: Where America’s Broadband Networks Really Stand, and a number of blog posts responding to the original Cost of Connectivity report in 2012. However, their report update makes the same claims as the last one using the same logic: we therefore feel compelled to issue a very similar rebuttal. The NAF report fails to engage the issue in a way that helps us understand what is really going on in our broadband markets.
There are two important questions to consider when examining these broadband markets: are companies offering a fair price for high-quality broadband services given their costs, and are they competing in a way that will offer dynamic improvements in the future? The NAF report attempts to answer these questions implicitly, by showing costs in U.S. cities relative to cities abroad. Comparisons like these can be revealing, certainly, but they can also be misleading. If we are trying to use comparative data to understand whether U.S. broadband provision is offering competitive services at competitive prices, we need to understand first of all whether the services are comparable, and second of all whether the providers face similar costs. A third consideration is the dynamic component—are providers investing enough in our infrastructure to keep our broadband competitive in the future?
Apples to Apples?
One of the most striking problems with the NAF report is that it leads with its “Triple Play” comparison of bundled internet, television, and phone services. While it is not entirely nonsensical to compare these different services across countries, the fact is that customers in different countries will in fact receive substantially different bundles. Different licensing agreements in different countries mean that even basic TV bundles may cost different amounts. Also, many “basic” bundles in the U.S. (such as Verizon FIOS and all of the listed Comcast plans) include channels that would cost extra in other countries, particularly sports channels like ESPN. And while internet service may be relatively comparable between countries (although not entirely, as we will discuss), phone service in the U.S. typically includes unlimited long distance within the U.S.
Internet speed is another important factor. As we have noted before, speed claims in Europe appear to be overstated, significantly more than they are in the United States. This shouldn’t take us by surprise, since we know that the U.S. has faster broadband infrastructure—a claim that is borne out by Akamai and OECD deployment data.
Finally, pricing is a complex topic that the NAF report attempts to simplify but does not do a great job of. Although they have the data, the NAF study does not appear to take into account activation fees, modem rental, or other fees. For example, VTX in Zurich has a $5.76 per month modem rental fee that is not included in their $29.96 monthly price. It seems odd that having this data, they did not opt to include it in their ranking. Moreover, while it may make sense not to take promotional pricing into account, it does distort the rate that customers may actually pay over the short or even long term.
The limited reliability of the NAF pricing data can be seen clearly when we compare it to other studies using different methodology. Most interesting is the 2012 International Telecommunication Union (ITU) study, Measuring the Information Society, which ranks the U.S. 3rd in the world for cheapest wired broadband. While the ITU study has a lower bar for what constitutes broadband than most current definitions, it has the advantage of using prices from the providers in a country that have the largest market share, thus providing a much more realistic picture of what subscribers are actually using.
It’s a Big Country
Even assuming that we are able to find broadband plans similar enough to compare between cities in different countries on different continents, that tells us nothing about whether the prices are fair. This is because broadband providers may face very different cost structures depending on their location. Businesses face a huge variety of potential costs, from R&D expenditures to legal fees to raw materials, but there are several areas where the costs for broadband providers particularly differ.
Taxes and subsidies present what are perhaps the most straightforward differences in broadband costs across locations and jurisdictions. Korea and Japan have historically had significant tax incentives for broadband infrastructure buildout, which eventually comes out of taxpayer wallets but not in a way that is easy to add into consumer price comparisons. Regarding taxes, the U.S. has had a particular disadvantage with respect to the phone-bundling portion of Triple Play packages in cities in the form of phone taxes. U.S. phone providers have to pay an array of fees and taxes like the Spanish-American War Tax and a Universal Service Fund contribution to cover service in the countryside at the same prices that city dwellers pay.
The prices we pay for telephone service in the United States reflect our historic commitment to uniform pricing of communications services across the nation despite the varying costs of providing service according to population density and distance. So even a cursory examination finds that most broadband services provided by cable companies or traditional phone companies are priced the same in hard-to-serve areas of low population density and cheap-to-serve high-rise apartments in the big cities. Broadband service doesn’t have to be priced this way, of course, but our carriers have always done it that way with a few exceptions in areas where they’re especially eager to attract new customers or to retain old ones.
In contrast, many of the low-priced foreign and domestic competitors NAF features serve almost exclusively high-density, urban areas. It’s much less expensive to serve urban customers, so firms that specialize in urban markets can naturally price their services lower than large American ISPs that must serve urban, suburban and rural customers. The effect of this study’s focus on city-to-city comparisons is to distort the playing field by comparing the prices charged in low-cost and high-cost scenarios as if they were the same.
The Universal Service Tax and our commitment to providing equal service thus functions as a tax for the cities, but for rural areas it may in fact function as a subsidy. It is one way that we attempt to deal with the infrastructure challenges that have been borne out of our low population density. Low population density is a problem because it makes providing infrastructure much more expensive, and infrastructure occupies a significant portion of most telecom companies’ budgets.
Because different companies face radically different infrastructure costs depending on their location, comparisons between provider prices can quickly become incoherent. Broadband has high fixed costs (stringing cable, setting up home connections, installing neighborhood boxes, etc.), and those costs can change dramatically depending on location. This is further complicated by the fact that telecom companies may in fact be selective about the clients they serve. For example, the NAF report includes Webpass, a San-Francisco-based provider that only serves large apartment buildings. The costs facing Webpass to connect an additional apartment to fiberoptic cable are a far cry from the costs facing Verizon to connect a rural farmer in Nebraska.
The infrastructure cost problem biases the entire report against U.S. cities because their population density is far lower than their European or Asian counterparts. Since infrastructure costs in the U.S. are naturally going to be higher than those abroad, direct price comparisons will not provide real data on whether higher prices for broadband and other services are fair.
While assessing relative price versus cost is problematic, we can make some educated guesses by comparing U.S. companies to their foreign counterparts as well as local municipal providers. Telecom companies in the U.S. appear to have significantly leaner profit margins than those in Europe, as we’ve mentioned before in our broadband report. Moreover, municipal broadband networks like those in Chattanooga or Lafayette end up charging prices similar to those of private providers. While municipal providers have in some cases undercut private providers and pushed them to lower their rates, this is due as much to cheaper financing costs and more densely populated regions, as it is to corporate profits.
And The Future
The final question is whether we are being too complacent and whether we need to take policy action to beef up our broadband or bring down the price. Given that nationwide our low-priced broadband is actually some of the most affordable (according to the ITU report) and that we are holding our own on speed (according to recent Akamai speed rankings), panic seems unwarranted.
The one place we might be worried about is the affordability of high-speed services. We are likely to see large gains in these areas in the future, however, as a payoff for our recent buildout of fiber-optic cables. After overbuilding during the 1990s tech boom, and then a lull, the United States has recovered and put in much more fiber than Europe in recent years, giving us faster and more reliable plans to choose from.
U.S. broadband competitiveness is important, and we must always be wary of powerful market players. But simplistic comparisons that gloss over complicated fee and plan structures and make no effort to examine actual costs to providers are not helping us formulate better policy—instead, by raising alarm where none is due, they prevent us from thinking clearly about what our broadband markets actually look like.
While we understand that the data collection process is a difficult one, some of the NAF data does seem to be suspect. EE (London) does not appear to have 100mbps speeds for their mobile dongles (see this article from April, trumpeting 20 mbps instead).