I am American earning an industrial PhD in internet economics in Denmark, one of the countries that law professor Susan Crawford praises in her book Captive Audience: The Telecom Industry and Monopoly Power in the New Gilded Age. The crise du jour in America today is broadband, and Susan Crawford is echoed by journalists David Carr, John Judis and Eduardo Porter and publications such as the New York Times, New Republic, Wired, Bloomberg News, and Huffington Post. One can also read David Cay Johnston’s The Fine Print: How Big Companies Use ‘Plain English’ to Rob You Blind.
It has become fashionable to write that American broadband internet is slow and expensive and that cable and telecom companies are holding back the future—even though the data shows otherwise. We can count on the ”America is falling behind” genre of business literature to keep us in a state of alert while it ensures a steady stream of book sales and traffic to news websites.
After six months of pro-Crawford coverage, the New York Times finally published two op-eds which offered a counter view to the “America is falling behind in broadband” mantra. Crawford complained about this in Salon.com and posted a 23 page blog on the Roosevelt Institute website to present “the facts”, but she didn’t mention that the New York Times printed two of her op-eds and featured her in two interviews for promotion of her book. I read Crawford’s book closely as well as her long blog post, including the the references she provides. I address Crawford’s charges as questions in four blogs.
- Do Europeans and East Asians have better and cheaper broadband than Americans?
- Is fiber to the home the network of the future (FTTH), or are there competing technologies?
- Is there really a cable/mobile duopoly in broadband?
- What is the #1 reason why older Americans use the internet?
How the broadband myth got started
Crawford’s book quotes a statistic from Akamai in 2009. That year was the nadir of the average measured connection speed for the US, placing it at #22 and falling. Certainly presenting the number at its worse point strengthens Crawford’s case for slow speeds. However, Akamai’s State of the Internet Report is released quarterly, so there should have been no problem for Crawford to include a more recent figure in time for her book’s publication in December 2012. Presently the US ranks #9 for the same measure. Clearly the US is not falling behind if its ranking on average measured speed steadily increased from 22nd to 9th.
Crawford notes on her blog ”Tussling over contestable rankings is not a good use of our time” and then proceeds to list the rankings of the US from a number of content delivery networks. She does not explain, however, the implication of this measurement. Akamai is the world’s largest content delivery network, speeding over one-third of all the content on the web and capturing 1 billion IP addresses per day. It is the most reliable longitudinal measure, but its methodology should be clarified.
Akamai measures speeds in a similar way to how cars are clocked on a freeway. For example a radar detector can measure the speed of a car at any moment, say 50 mph. However that car could go 100mph or 25 mph. It’s just what is captured at the moment of measurement. As for broadband, there may be a 100 Mpbs connection to a person’s home, but if that subscriber only signs up for 5 Mbps, Akamai will only report 5 Mbps. As a matter of fact the Akamai Q1 2013 report shows Washington DC, Vermont and New Jersey with higher average peak speeds than South Korea, the #1 country. For an in depth discussion of broadband statistics see The Whole Picture: Where America’s Broadband Networks Really Stand from the Information Technology & Innovation Forum.
Incidentally recent reports from both the Federal Communications Commission and the White House (Office of Science & Technology Policy and the National Economic Council) contradict the dour picture critics paint about American broadband. See the report Four Years of Broadband Growth.
Don’t Europeans and East Asians have better and cheaper broadband than Americans?
This is the wrong question. The question we should ask ourselves instead is how well have nations applied broadband technologies to improve their economy and standard of living? With all this discussion about speed, some consider ultra-high-speed wired broadband for its own sake, as an end in itself. But bandwidth alone does not an economy make. Instead we need to envision broadband as in important input to the information economy ecosystem.
Citing a report from the New America Foundation, Crawford asserts that Americans pay ”three or four times” more for the same download services as in other countries. The fact of the matter is that I can find broadband prices both higher and lower around the world. The website of the leading Danish broadband provider TDC offers a package of 24 channels, 20Mbps broadband, and either fixed telephony or 4 hours mobile telephony for for 414 DKK ($58.73 + 25% tax = $73.41). There is a one-time fee of 399 DKK ($70 + 25% tax=$88).The similar monthly package goes for $60-$70 in the US. The next level package of 50 Mbps is $80. So in this example, broadband is only slightly more expensive, depending on the local tax, than in the US. Indeed the OECD points out Spain and Norway as some of the most expensive countries for broadband. Keep in mind as well that most of Denmark’s residents live in the major cities in apartments or in houses more closely packed that a typical American suburb, which also explains some of the price difference.
To be sure, we can find countries where broadband may be less expensive, but gasoline is four times as high. Local conditions and taxation will change the price. For this reason economists use a basket of goods and services to when evaluating consumer prices. The market price of broadband in two countries may not reflect the same inputs. The price can vary for many reasons including the network type, the network speed, the type of subscriber (individual, business, company etc), whether the item is sold in bundle, whether the subscriber has a certain exemption, taxes, and other factors not limited to geography, density and so on. Economists and financial analysts who study prices build complex, dynamic models to reflect these factors.
The OECD provides the most comprehensive, global information on broadband prices, but it relies of national governments to provide the data, rather than collecting it directly from retailers or websites. This challenge to determine “the facts” is also exacerbated by competing sources. Indeed the most comprehensive source of broadband information comes from the Organization of Economic Cooperation and Development (OECD). Their Broadband Portal offers a wealth of data on many broadband measurements. Plus the new OECD Communications Outlook published in July 2013, has the most recent comparison for prices globally.
Other positive information about the US appears in the the most recent from OECD report. It notes the decline in the price of megabit per second of advertised speed. In the 2011 report showed that the US ranged between $1.10-71.49, but that number has fallen to $0.53-$41.70 in the 2013 report. That translates to a 51% improvement at the low end and a 41% improvement at the high end. Some countries have lower prices, but the decline of the price for the US shows that thing are getting better, not worse, for broadband.
Let’s look at the mobile example. Using information from Bernstein Research and analyst Craig Moffett, Crawford asserts that mobile prices are too high. On this point, one should defer to the GSM, the global standards organization for the mobile industry. Its report on mobile in the US and Europe notes that yes, Americans do pay more for mobile than Europeans ($69 vs. $38 for an average monthly subscription), but Americans use five times more voice and twice as much data. From a G-20 perspective the OECD notes that “Given that mobile broadband constitutes a relatively new market compared to fixed broadband, there tends to be greater experimentation in wireless markets. Moreover, the evolution of the smartphone ecosystem has resulted in a complex array of stakeholders who determine these prices.“ 
Furthermore Europeans may have lower prices for mobile, but this is because wholesale rates are regulated to be artificially low. European consumers have a low price in the short term, but in the long run they are shortchanged because European carriers haven’t made enough profits to invest in infrastructure. This is outcome of the “services based competition model” (allowing new entrants to resell incumbent’s services at a low price) which hasn’t panned out to deliver the infrastructure investments as hoped. Clinton Administration economist Ev Ehrlich who also studies this issue published his op-ed The Myth of America’s Inferior Broadband in the Wall Street Journal describing this situation.
Finally, roaming prices are still not harmonized, so when a European travels from one country to another, there are surcharges on calls and SMS. Imagine if you were charged a different rate each time you entered a new state in the US. Such is the case in Europe.
OECD updated pricing information published in July 2013 notes that entry level prices should be no more than $30 purchasing power parity. The America entry level monthly price is $27.Thereafter, if people want faster speeds, they pay for it. That is only fair. This means that for as little as $27, people can be assured bandwidth to do essential email and web browsing for job applications, online banking, and so forth. A forthcoming blog post will investigate the issue of low-income Americans for whom $27 is too much.
As for people who pay $100 or more per month for broadband, which amounts to cost of a daily visit to Starbucks, that number should be put into perspective by measuring it against the cost to purchase the same content and communication services piecemeal. One would have to add up the price of all the newspaper subscriptions, the movie tickets, the DVDs, the CDs, the long distance calls as well as to add some kind of premium to cover the applications we have today that never existed before the web. On balance, broadband is a tremendous value in the US. When one sees cheaper prices in other countries, one needs to consider that often these citizens pay three times for broadband: with subscriptions, with rent/home owners fees, and with taxes. The benefit for Americans is that they pay for broadband once, and they pay what it costs.
High Speed Broadband Adoption
From the research perspective, the leaders of my institute, The Center for Communication, Media and Information Studies in Copenhagen, published a report titled “Broadband Bandwidths in a 2020 Perspective” reflecting on the needs of developed countries such as the US. Their assessment is speeds are increasing faster than consumers demand them. The report notes that in Denmark, one of the perennial top performing countries in the OECD for broadband adoption, 65% of homes are passed by a broadband technology that can deliver 100 Mbps, but only 0.7% subscribe to the fastest tier. Danes can get what they need from lesser speeds, and the price of the faster service is not justified from their perspective.
Even during the financial crisis in 2009, the supposed low point of performance of broadband speed, activity on America’s broadband networks was in full swing. At the time I worked for a web analytics software company in Silicon Valley. Our software was enabled on over 2000 enterprise websites visited by millions of Americans every day. These websites ran the gamut: ecommerce, news, banking, education, student financial aid applications, B2B, video-embedded media, and so on. Never once did our customers complain that there a was not sufficient broadband for end users’ needs, that they were missing out on customers because of lack of broadband access, or that speeds were too slow. On the contrary, broadband had enabled new markets. Broadband was so ubiquitous that it was no longer a differentiator. Thus these companies wanted to deploy every additional advantage, including search marketing, behavioral targeting, multivariate testing and so on. Even more impressive was that almost none of these companies were based in major cities or even Silicon Valley. In that way, broadband brought the death of distance.
Broadband and Employment
We can learn a lot of the folly of the broadband for its own sake mentality from South Korea, #1 in Akamai’s study with an average measured speed of 45 Mbps. Their primary uses of broadband are by far video game entertainment for consumers and video conferencing for businesses. The problem with these two applications is that they drive little revenue versus the traffic they consume on the web. Much of real time entertainment is piracy, and the money in games is largely in the hardware. As for online gaming, less than 5% of players pay for games. Video conferencing was thought to be a great revenue opportunity for platform providers, but users are choosing free versions of Skype instead. So these two endeavors don’t generate the cash flow that create jobs.
Broadband has enabled some industrial productivity and supports a marginal “Gangam Style” entertainment economy in South Korea. It is estimated that performer Psy made about $8 million from his famous song, including the 1.6 billion YouTube views and the iTunes sales. Few performers will ever achieve that level of success. His is not a replicable business model, let alone a business case for broadband. The real money in South Korea’s economy still comes from electronics, automobiles, shipbuilding, semiconductors, steel, and chemicals — the same growth engines from the pre-broadband days. Ditto for Japan and Sweden.
Most important, the national broadband project in South Korea has not yielded the jobs that were expected. Broadband has enabled entertainment but not employment. A new report by the Korea Information Society Development Institute, “A Study on the Impact of New ICT Service and Technology on Employment,” bemoans the situation of “jobless growth.” The government is also concerned about internet addiction, which afflicts some 10 percent of the country’s children aged between 10 and 19, who essentially function only for online gaming but not in other areas of society.
Europe also has challenges translating broadband into employment. My colleague at the Ifo Center for the Economics of Education and Innovation in Munich published the results of her econometric study of the impact of broadband internet on employment on 8460 municipalities in West Germany. Over the last five years the German government has invested €454 million (almost $600 million) to bring broadband to the rural areas. Though there is an impact on local employment by local broadband infrastructure, the impact is very slight. The econometric study shows that an increase of DSL penetration by 10% yields between 0.03-0.16% increase in employment. This research suggests is that broadband alone is not enough to stimulate employment. Other factors such as level of education, professional skills, existing employment opportunities, types of extant industries and so on also play a role in employment.
Broadband and Economic Growth
What is important about broadband is not measuring speeds and counting rank, but turning technology in productive use in the economy. In spite of all of the challenges of broadband, America leads the world in broadband-based industries. Mary Meeker of Kleiner Perkins Caulfield Byers assessed the world’s top internet companies, and found the US an unusually strong performer. Of the top 25, the US had the most, 14; China, 3; Japan, 2; South Korea, 2; Russia, 2; and the UK and Argentina each have 1. The point is that the USA, with just a fraction of the world’s internet users and with an oversized investment (one-quarter of the world’s financial outlay in internet infrastructure) has been able to leverage broadband into over $1 trillion of market value in 2013 alone with just 14 companies. This is a stunning achievement, and it does not even take into account all of the small and medium sized American companies that would have never existed without broadband.
Given Crawford’s supposition of alleged high prices that limit adoption and force consumers to slower speeds, I conclude the opposite after reviewing the data. Consumers have broadband at all price levels as well as speeds. The vast economic growth and the transformation of the US from and industrial to an informational economy means that the US gets a lot of bang for its broadband buck. Whatever the circumstances, the US has managed to turn broadband into productive use better than other nations. To be sure, the broadband and economic development equation is complex, but it’s not true that Europeans and East Asians have it better when it comes to broadband. The next blog post addresses the question of whether fiber to the home will be the network of the future or whether network technologies will compete.