Skip to content Skip to navigation

Household Demand for Broadband Internet Service

Feb 2010
Working Paper
09-008
By  Gregory Rosston, Scott Savage, Donald Waldman
As part of the Federal Communications Commission (“FCC”) National Broadband Report to Congress, we have been asked to conduct a survey to help determine consumer valuations of different aspects of broadband Internet service. This report details our methodology, sample and preliminary results. We do not provide policy recommendations. This draft report uses data obtained from a nationwide survey during late December 2009 and early January 2010 to estimate household demand for broadband Internet service. The report combines household data, obtained from choices in a real market and an experimental setting, with a discrete-choice model to estimate the marginal willingness-to-pay (WTP) for improvements in eight Internet service characteristics. The first three are standard features for all current Internet services and include: cost; connection speed; and the reliability of the connection to the Internet. The remaining five characteristics are new activities that could be bundled with future Internet services. They include the ability to connect to the Internet wirelessly from outside the home, download and watch high-definition movies, designate certain downloads as high-priority, interact with health specialists, and place free videophone calls over the Internet. Choice experiments are used to estimate household preferences. Respondents are presented with eight choice scenarios, and in each scenario, must choose between a pair of Internet service alternatives that differ by the levels of their characteristics. The information in these choices is enriched with market data by having respondents indicate whether they would stay with their current (actual) Internet service or switch to the hypothetical service they had just selected. The marginal utility parameters of the representative household's utility function, and WTP, are then estimated from all observed choices. (Continued)