Public finance economics is often concerned with the "unintended consequences" of government policies, as an evaluation of these unintended effects is required to obtain a complete picture of the efficacy of a given policy. In my dissertation, I study three government policies, along with their potential unintended effects. I find that, in general, the potential unintended effects that I study are economically small. These findings suggest that policymakers in these areas can focus relatively more on the direct, intended effects of these policies when evaluating their costs and benefits.

In the first chapter, I study an expansion of Medicaid which occurred in 2014 as part of the Affordable Care Act. Because this expansion was only taken up by approximately half of all U.S. states, migration across state lines is a potential unintended consequence of these reforms. I analyze data from the American Community Survey and find that this migration response was not large: I estimate that migration could have increased Medicaid rolls in expansion states by no more than 2 percent.

In the second chapter, Chris Boone, Arindrajit Dube, Ethan Kaplan, and I study extensions of unemployment insurance (UI) in the U.S. during the Great Recession. The "moral hazard" effect of UI on individual search effort has been well-studied. In this chapter, we broaden these potential unintended effects to include all potential effects (positive and negative) on aggregate employment at the county level. Again, we find little effect: we can rule out negative effects of the extensions in excess of -0.3 percentage points of the employment-to-population ratio.

Finally, in the third chapter, I use administrative tax data to study a tax reform in the state of Kansas which was designed to provide tax relief to businesses. This reform also increased an incentive for owners of a subset of these businesses to reclassify wage income as the profits of the firm -- potentially reducing tax revenue. I again find that the effect was small. I show theoretically why this increase in incentive may not have caused a large response, even if the introduction of the incentive caused a much bigger response.