Abstract

The Federal Communication Commission reports that, as of June 2006, broadband service is available in 99% of all ZIP Codes in the United States and to the population living within them. Out of the 64.6 million high-speed Internet lines available in the United States, 50.3 million were designed primarily to serve residential end users. Are these optimistic figures the right proxy of actual use? In addition, to what extent did the narrowing gap between DSL and cable modem contribute to the growth of broadband penetration?

This thesis examines median household income and the number of broadband providers and their relationship to broadband penetration in two sets of metropolitan areas, i.e. the top 30 and bottom 30 metropolitan areas in terms of household broadband subscribership. It finds that, while income and the number of service providers do not necessarily lead to a higher broadband penetration rate in a metropolitan area, higher income and more providers appear in the top 30 areas.

It finds no evidence that the broadband growth was led mainly by DSL. Rather, local characteristics, industry for instance, play an important role in some of the areas with higher broadband penetration. Contrasting the top 30 to the bottom 30, this thesis also finds that dial-up can be a substitute for broadband in the latter group for connections with the Internet, and suggests ways to absorb dial-up user households to broadband in relation to the issue of digital divide.