Dollars, Cents, and Megabits: An Analysis of State Broadband Programs

Abstract

With billions of federal dollars pouring into state broadband expansion programs, state efforts to bring the modern internet to all Americans have been thrust to the fore. Little research into the effectiveness of these programs has been done, however, and many programs are either brand new or only a few years old. This thesis attempts to determine how effective these existing programs have been at efficiently spending taxpayer money. It uses both quantitative and qualitative methods. It also draws upon a detailed data set of nearly one thousand state grant projects gathered from broadband agency websites and public records requests. The thesis first looked at unnecessary grants, those grants where the state provided government funding to a broadband expansion project that would have gone forward without government intervention. Our thesis used a novel economic model of the net present value of a broadband project to determine which grants were unnecessary and provided evidence that some states might have high levels of unnecessary grants. An analysis of case studies of unnecessary grants confirmed their existence but also showed some of the flaws in our model and that not every grant it marked as unnecessary was actually so. Next the thesis examined grants where state governments paid significantly more per premises than the federal government did in similar programs. Here, a series of Welch’s t-tests found statistically significant evidence that some states had a higher average grant cost per premises compared to the federal government. By studying some examples we were able to determine some possible explanations for this behavior. Finally, we looked at grants where alternative technology, like satellite broadband, would cost less to meet the states’ goals than the funded project. While we found many grants with project costs per premises that exceeded those of satellite, interviews with state officials revealed that this was because states have not yet received any applications using alternative technology. The conclusions drawn by this thesis could be useful to states as they define policies for their new broadband programs and should guide further research in the area.