The Stagnant Price of Bandwidth

Title: Evidence of a Modest Price Decline in US Broadband Services (Subscription or fee required.)

Authors: Shane Greenstein and Ryan McDevitt (both affiliated with Northwestern University)

Publisher: The National Bureau of Economic Research, Working Paper No. 16166

Date Published: July 2010

Between 2000 and 2009, the percentage of households with broadband Internet connections in the U.S. grew to 63.5 percent, or more than 70 million households, from 4.4 percent, according to the Pew Internet & American Life Project survey. By 2006, the last year for which U.S. census data is available, U.S. households were spending US$39 billion a year on Internet access, with 72 percent of that money going toward broadband access. The authors of this paper set out to find why that phenomenal growth in usage didn’t correspond with a significant drop in the cost of broadband service, and what impact stubbornly high broadband prices may have on government policy and industry innovation.

The authors note that it is common in consumer electronics markets for prices to fall dramatically, even as new technologies evolve and become more powerful. Personal computers, digital cameras, and computer chips all experienced yearly price declines exceeding double digits during the past decade, even as the underlying technology improved. In contrast, after analyzing 1,500 contracts of DSL and cable service companies from 2004 to 2009, the authors found that the price of broadband declined only nominally, between 3 and 10 percent. The study looked both at consumers who purchased broadband Internet access as a stand-alone service, as well as those who signed up for bundled agreements, in which Internet access is provided along with television or Web-based phone service (Comcast’s Triple Play, for example).

The authors found that although prices were flat, customers got a bit more for their money as the technology matured. For example, in 2004, the median cable modem contract price was $45 per month for an upload bandwidth of 3000 bits per second (bps); in 2009 it cost $53 per month for 8000 bps. Although U.S. policymakers believed that the widespread adoption of the technology would contribute to economic growth, which it undoubtedly did, the authors argue that government decisions to deregulate the telecommunications market led to the relatively static prices over the last five years. Only a handful of companies could afford the massive infrastructure costs associated with building communications networks, and without rules in place to force these large firms to open their networks to competitors, deregulation actually discouraged competition in the market. With little competition, there has been no incentive to lower prices or to make significant upgrades, the authors contend. They recommend that U.S. regulators create a price index to track the movement of broadband prices over time and determine whether prices are artificially high.

Bottom Line:
Unlike the price trajectory of other technologies, the price for broadband Internet access has declined only slightly during broadband’s period of widespread adoption. The likely reason, the authors conclude, is that the unique nature of the broadband market, and the particular path of its development, has led to an unusual lack of competition.