UC Berkeley

President Obama -- in the executive order Using Behavioral Science Insights to Better Serve the American People -- states that: ``To more fully realize the benefits of behavioral insights and deliver better results at a lower cost for the American people, the Federal Government should design its policies and programs to reflect our best understanding of how people engage with, participate in, use, and respond to those policies and programs.'' The conventional assumption is that individuals respond rationally to incentives. Most government programs have been designed on this premise. This dissertation aims at empirically analyzing the responses of individuals to government programs in light of the findings from behavioral sciences in order to design better interventions.

One of the most important interaction individuals and firms have with the government is through the tax system. Virtually every transaction is subject to taxes, be it consumption taxes, income taxes, property taxes, etc. As such, taxes constantly affect our behavior in directions that are not always properly understood.

In the first chapter, I show that individuals forgo substantial tax benefits to avoid the hassle costs of filling out forms and collecting receipts. To do so, I use a quasi-experimental design and a novel identification strategy. Employing a sample of US income tax returns, I observe the preferences of taxpayers when choosing between itemizing deductions and claiming the standard deduction. Taxpayers forgo tax savings to avoid the hassle cost of itemizing, resulting in an average burden of itemizing of $644, with substantial heterogeneity. A revealed preference argument implies that itemizing deductions is as painful as working 19 hours. The burden of tax filing is larger for richer households, consistent with the fact that the value of time increases with income.

The second chapter explores two explanations for the large magnitude of forgone deductions. First, it could be due to an extreme aversion to filing taxes. Such aversion implies that itemizing deductions imposes aggregate hassle costs of 0.2% of GDP and back-of-the-envelope extrapolations to filing federal taxes yield an overall burden of 1.25% of GDP. Second, if taxpayers are time inconsistent they may forgo large benefits even when hassle costs are relatively small due to procrastination. I provide evidence most consistent with taxpayers being present-biased. Both explanations -- whether driven by preferences or mistakes -- suggest that the burden of tax filing is significantly larger than previously estimated. I also discuss policy implications of the result in light of each explanation.

In the third chapter I document the existence of a novel dataset that can be used to study the international migration of high skilled workers. In particular, this type of datasets can inform researchers and policymakers on the extent to which taxpayers migrate to avoid taxes. I provide new series on the international migration of high skilled workers educated in France from 1944 to 2012. To do so, I use alumni databases to track the location of graduates from leading French post-secondary institutions. The proportion of high skilled individuals working in France has been steadily decreasing from 1944 to 2004. Recent years have seen an increase in the percentage of graduates staying in France in contradiction with the view that high taxes and administrative costs have been leading high skilled workers to leave France.