The Federal Reserve Bank of New York has an interesting report on broadband affordability…

Existing data on broadband pricing is often limited to aggregate price estimates at the state level or for metropolitan areas. The FCC Urban Rate Survey, an annual survey of the fixed voice and broadband service rates offered to consumers in urban areas, aims to provide reasonable comparability benchmarks for fixed voice and broadband rates for universal service purposes.29 The
data provides information about providers in each state and the average rates offered for different broadband technologies, plans, and speeds, but does not offer information about serviced geographies within each state for each provider.30
This report uses the FCC Urban Rate Survey to illustrate associations between broadband speeds, technology type, and price. However, the broader analysis utilizes data from The Markup’s point-in time estimates of address-level broadband pricing data from cities,31 the “How We Uncovered Disparities in Internet Deals” dataset used to conduct a report on disparities in Internet speeds
offered for the same price in low-income versus middle-and-high income neighborhoods32 The data, collected in 2022, uses information from 800,000 Internet plans from four Internet service providers. Pricing information from these plans, representing the lowest price offered by each plan, is available at the address level for 38 cities and provides information about average download and
upload speeds offered for each plan.
There are limitations that come with using the Internet plan pricing data provided by The Markup. One, data is sampled from only one year (2022) and therefore cannot be used to illustrate changes in price over time. Two, the data often only contains pricing information for one or two providers per city, a phenomenon which is consistent with the way internet service providers often cover distinct geographies, however, it reduces variation in the data. The data, as it is used in this study, is only intended to provide illustrative, quantitative information on what broadband prices look like relative to median household income in cities but may or may not be representative.

It’s an interesting concept and I’d love to see a rural version of it. We don’t all define affordable by the same dollar amount (for any purchase!) so percentage of income makes sense. But the broadband services are not all the same and availability isn’t the same. What’s the cost of needing one service for speed and another for reliability? Or the cost of having no choices? These are factors that seem to be more prevalent in rural areas.