A new measure of US residential broadband availability

The only comprehensive published indicator of residential broadband availability in the US is number of providers in each ZIP code, as reported by the Federal Communications Commission (FCC). This measure has been widely used in academic and policy research to assess availability and to identify under-served areas, but it is acknowledged to be flawed and is often misinterpreted. This paper develops an alternative measure of residential broadband availability. Using the December 2005 FCC data and individual broadband adoption data from Forrester Research, the paper estimates a relationship between the number of providers in a ZIP code and the level of residential broadband availability in the ZIP code. Broadband is estimated to be available to 53% of households in ZIP codes with 1-3 providers, rising to 100% in ZIP codes with 14 or more providers. Aggregating these estimates at the national level implies that broadband was available to 85% of households in December 2005. Availability was highest in southern and western metropolitan areas like Miami and San Jose. Using these estimates of availability, population density and average income both have positive and highly statistically significant effects on broadband supply. The results provide a user-friendly tool to help policymakers assess broadband availability. The estimates are also useful for future research about the effects of broadband availability. (C) 2009 Elsevier Ltd. All rights reserved.