Essays in Public Finance

Thumbnail Image



Journal Title

Journal ISSN

Volume Title

Repository Usage Stats



This dissertation addresses three interrelated questions regarding the markets through which governments raise funds. Public entities around the world commonly raise revenue by levying taxes or by selling bonds. I examine aspects of both mechanisms in the US context as well as their interaction. Focusing on how public financial decisions affect private markets in the US, I examine (1) the process by which municipal bonds are created and sold, (2) how tax treatment of municipal bond income affects the underwriting process, and (3) how corporate tax incentives change worker outcomes.

Like in many financial markets, the governments issuing municipal bonds often don't have the expertise to accurately construct debt packages, price individual bonds, or place bonds with final investors. To interact with the market, municipalities often hire two types of intermediaries: financial advisors and underwriters. The financial advisors structure a bond package while the underwriters buy these bonds from municipalities and sell them to investors. The second chapter examines a policy change from the Dodd-Frank Wall Street Reform and Consumer Protection Act of 2010 that requires these two distinct roles are filled by different firms. To determine if allowing both underwriter and advisor to be the same firm distorts outcomes, I use a difference in differences approach to measure bond outcomes for municipalities who hired different types of advisors before and after the Dodd-Frank regulation. I find bonds sold with an advisor who could also be an underwriter experience a decrease in interest costs after the advisor is no longer allowed to underwrite. The improvement in outcomes from the municipality's perspective is driven in part by a large increase in competition from non-advisor underwriters that more than offsets missing competition from the advisor--consistent with the partial resolution of a winner's curse.

The third chapter also sheds light on how the imperfect competition in the underwriting market for municipal bonds affects borrowing outcomes. Most municipal bonds sold in the US pay interest that is exempt from personal income taxation at federal and state levels. This personal income tax exemption is contentious because, in isolation, it primarily benefits individuals paying top marginal tax rates. Using variation in federal taxes, state taxes, and their interaction from 2008 to 2015, I first show that increases in taxes are associated with surprisingly large decreases in borrowing costs for municipalities. This fact motivates the estimation of an empirical auction model with conditionally independent private values that features endogenous entry of potential bidders, endogenous entry into an auction stage, and unobserved heterogeneity across auctions. The model shows that very large elasticities of markups drive larger than unity passthrough from the tax advantages to municipal borrowing costs on average. The model is used to estimate counterfactual tax regimes and highlights how changes in federal tax rules have heterogeneous effects across space with some states gaining and others losing with the Tax Cuts and Jobs Act of 2017.

The fourth chapter empirically measures how another federal tax incentive, ``Bonus'' depreciation, affects areas across the US differently. Incentives for capital investment are used in the US and around the world to inspire business growth with the stated goals of creating jobs. Using a panel regression model and cross sectional variation in exposure to the types of firms that benefit from Bonus depreciation, I show that moving from the 25th percentile to the 75th percentile of exposure to treated firms increases local employment by about 3%.






Garrett, Daniel G (2021). Essays in Public Finance. Dissertation, Duke University. Retrieved from


Dukes student scholarship is made available to the public using a Creative Commons Attribution / Non-commercial / No derivative (CC-BY-NC-ND) license.