The Economics of Universal Service: An Analysis of Entry Subsidies for Rural Broadband

Andre Boik is an assistant professor of economics. His project, which explores the use of public funds to subsidize rural high speed wired broadband, was awarded an ISS Individual Research Grant in 2016. He provided this update in June 2017.

What motivated you to pursue this project?

I became interested in the topic after learning about a government project called the Connect America Fund which uses billions of dollars of public funds to subsidize private broadband firms to enter into areas which currently do not have access to high speed wired broadband.  The source of these public funds are “universal service” fees that most consumers would recognize from their telephone bills. These funds were originally used to subsidize the costs of rural telephone providers, but are now being transitioned to subsidize rural high speed wired broadband. I wondered whether this was a good use of public funds since many rural areas currently have access to other forms of internet access: satellite, DSL, fixed wireless, etc. 

How has it progressed since you received an ISS Individual Research Grant?

In the paper, I estimate how much rural households value high speed broadband compared to the alternatives that already exist in rural areas. To do so, I use household-level data from North Carolina which involves over 4 million households. Combining several different sources of data was not an easy task in this context, since not all sources use the same address format. This has since been completed, the analysis conducted, and the paper, "The economics of universal service: An analysis of entry subsidies for high speed broadband", was published in Information Economics & Policy in April 2017.

What notable or surprising findings can you share at this point?

I find evidence that rural households do not in general value high speed wired broadband enough to warrant entry subsidies into all rural areas of North Carolina. To be specific, I estimate that fewer than half of all rural households would even purchase it if it became available. The thinking is as follows: if entry subsidies are used to incentivize a private broadband provider to enter into a rural area that currently does not have such access, then afterwards there will be an environment where one private broadband provider is available. But there are such areas that look like that already in existence, and we can learn by looking at how many households adopt high speed wired broadband there. What I find is that in areas where high speed wired broadband is available, many households choose (slower) alternatives like DSL and satellite broadband. 

If households in areas where wired broadband is already available choose other, slower alternatives, we can reasonably expect that the same would occur in areas that obtain access to a wired broadband provider as a result of entry subsidies funded by universal service fees. After accounting for this, I estimate how much households would have to value broadband in these areas to justify various coverage objectives. For example, if the government objective is to reduce the number of rural areas without wired broadband by 50%, the required monthly subsidy per household that purchased broadband would be $177. If the government target is instead 95%, the required monthly subsidy would be $1519. What this suggests is that increasing the broadband footprint would become increasingly expensive, and policymakers will have to decide how far they want to go.

Learn more about Andre Boik.