Three Essays in State and Local Public Finance

This dissertation is a collection of essays that address policy-relevant issues in state and local taxation. In the first, I use property value and flood data from all class 1 residential properties in New York City to provide causal evidence of the effect of hurricane Sandy on assessed values. I use a two-way fixed effects framework to measure changes in assessed value inside and outside of the flood zone following the hurricane. I find that assessed values decline by 10.3 percent in the flood zone in the 10 years following the storm. This reduction in assessed value among flooded properties indicates that there was a shift in the tax burden away from high-risk properties, which may lessen the impact of future storms and/or sea-level rise.

In the second essay, in collaboration with Don Bruce and Bill Fox, we provide a detailed discussion of state and local sales tax features and the extent to which they have fostered sales tax competition in recent decades. Most notably, the 2018 supreme court case Wayfair v. South Dakota greatly improved state governments’ ability to enforce collection of sales taxes on a destination basis. We explore how greater destination taxation has influenced the location of consumer purchases and business locations. We provide suggestive evidence that sales tax collections have grown more in rural Tennessee counties and less in Tennessee border counties since Wayfair. Additionally, we find that collections have grown more since Wayfair in North Carolina counties along the Tennessee border. Using state-level data, we show that business applications have grown at much faster rates after Wayfair in states with the highest sales tax rates. We attribute this to the removal of the significant disincentive to establish sales tax nexus that dominated the pre-Wayfair environment.

In the third essay, I use monthly foot-traffic data to study how consumers respond to state implementation of economic nexus following the Wayfair decision. Because states implemented economic nexus at different times, I employ a doubly robust difference-in-differences estimator that accommodates staggered treatment and recovers cohort-specific effects on the number of unique visitors to brick-and- mortar retailers. While the findings suggest that there was no impact on all in-person retail activity, there was a measurable increase in cross-border retail visits following economic nexus implementation. This pattern suggests a return to traditional tax- avoidance methods, with consumers crossing state borders to minimize their sales tax liability in the absence of a tax-free online option.