Impacts of the USDA Broadband Expansion Programs on Broadband Speeds in the United States

The USDA considers broadband access a “modern-day necessity”. However, 22.3% of Americans in rural areas lack high-speed broadband access. In this study, we analyze the impact of four broadband speed expansion programs across the United States. Between 2010 and 2020, the Telecom Infrastructure Borrowers program began offering loans to state and local governments and local organizations to expand broadband infrastructure to underserved telecommunication areas, rural areas, and towns with a population of less than 5000. The Community Connect grants that started in 2013 focused on improving infrastructure to increase broadband speeds in underserved and rural areas. The Farm Bill and the ReConnect program directed funds towards infrastructure expansions in underserved areas across the United States. We construct a unique census tract level panel for the years 2014-2021, with information on loans and grants received by rural areas for broadband expansion from the USDA Rural Development online web maps and FCC broadband speeds data on the census tract level. Then we use this data and a staggered difference in difference approach to identify the effects of funding access on broadband outcomes. Our results show that overall, these programs (primarily driven by the Infrastructure Borrowers) increased median download speeds by around 2 Mbps (or about 0.15 SD) per year. However, our results also suggest that the parallel trends assumption, which is necessary for a causal interpretation of DD estimates, does not hold for any of the programs. We then use a newly developed method that allows us to construct confidence intervals of treatment effects that are robust to the violation of the parallel trend assumption. Using these confidence intervals, we cannot reject the null hypothesis that the treatment effect is zero.