A TALE OF TWO INTERNETS: An Analysis of Exogenously Introduced Internet Access on Local Economic Outcomes

Abstract

This thesis seeks to answer two questions: Do increases in Internet access improve a local area’s economic outcomes? And does the elasticity of this increase in an area’s economic outcomes differ by income level? If American public policy is designed to improve economic outcomes, the answer to the first question seems to be yes. Since 2009, the Department of Agriculture has given $4.2 billion to subsidize Internet access through the American Recovery and Reinvestment Act (ARRA). On the local and state level, governments have invested millions on fiber optic networks to offer their own Internet. Yet, despite larger government involvement in its provision, few studies have empirically answered the first question about its effect in the affirmative, let alone addressed the latter. I attempt to contribute to the literature by showing causality in case studies where there was a large increase in Internet access exogenous to economic outcomes. In particular, I choose three examples: Electric Power Board in Chattanooga, TN, Pine Telephone Company in Choctaw Nation, OK and Google Fiber in Kansas City, KS and MO. With American Community Survey (ACS) data from December 2015, I can see changes at the censustract level. Using difference-in-differences and least squares regression, I test to see if increases in Internet access increased four types of economic outcomes: median income, mean income, population and house value. I conclude, based on the results of these statistical models, that increasing Internet access improves incomes in each case study. Although I saw some increase in house values, it was not consistent across all studies. I also find that the elasticity of this increase from access differs by income levels. My results are contradictory. For Electric Power Board, I find that low-income income areas saw outcomes grow, but for Pine Telephone Company, I find that only outcomes for high-income areas increase. I suggest that increases in areas’ incomes can justify government involvement in providing access. These results have both efficiency and distribution implications for public policy, as the benefits from Internet access are evident, but not uniform.