| This dissertation consists of three chapters that explore behavioral responses to taxation. The first two chapters are largely empirical, drawing on administrative tax data to study income reporting decisions and withdrawals from Individual Retirement Accounts (IRAs). The third chapter is an exploration of optimal tax theory when markets are imperfectly competitive and consumers do not maximize their own utility.;Understanding the way taxpayers respond to the tax code is critical for revenue and welfare analyses of taxation. One way taxpayers may respond is by bunching at kink points in the tax schedule to avoid high marginal tax rates. In the first chapter, I study this phenomenon using over 400 million federal individual income tax returns in the United States from 1996 to 2014, analyzing state and federal statutory kinks as well as effective kinks created by tax credits and phase-outs of deductions and exemptions. Though most kinks do not cause statistically discernible bunching, I find strong responses at other kinks. Consistent with prior research, I see bunching patterns grow over time at the first kink in the Earned Income Tax Credit (EITC) schedule. In addition, I present new evidence documenting (i) the emergence and rapid rise of bunching at the second EITC kink and the Child Tax Credit refundability plateau, (ii) strong responses to the temporary Making Work Pay Tax Credit, and (iii) weak responses at three statutory kinks. Though the self-employed bunch more, I find wage earners also respond in recent years. In general, substantial bunching responses occur only at kinks that maximize tax credits, and the strongest response occurs at the point in the schedule that maximizes credits net of taxes owed.;Another feature of the tax code that taxpayers may respond to is the requirement to withdraw funds from tax-preferred savings vehicles such as IRAs. These accounts are a substantial source of retirement savings for current retirees. In 2013, individuals age 60 or older held ;The underlying reason for studying behavioral responses to taxation is to inform better tax policy. A large literature studies optimal tax policy under a variety of theoretical assumptions. In the third chapter, I add to this literature, exploring internalities, which arise whenever a decision maker fails to maximize his own expected utility. This can happen due to, e.g., imperfect information, cognitive biases, or lack of willpower. I consider what the implications are for the optimal taxation of goods when consumers suffer from internalities. Prior research on this topic is limited to markets with perfect competition and ignores the incentive that firms have to de-bias consumers. I show that relaxing these assumptions changes the resulting conclusions. First, I analyze a market with imperfect competition, specifically Hotelling's model of monopolistic competition. I find that internality correction, even if costless, is not always desirable in such markets because firms change prices when the demand for their product changes, and these price changes can cause an inefficient reallocation of goods. Second, I analyze optimal taxation of internalities when firms have the ability to de-bias consumers. In such cases, the standard logic, which prescribes taxes equal to marginal internalities, does not apply. I find that, in general, market incentives attenuate the optimal internality tax (or subsidy) relative to a model with no firm de-biasing technology.;Index words: Asset decumulation, behavioral economics, behavioral responses to taxation, bunching, cognitive bias, Individual Retirement Accounts, internalities, internality taxes, investment, market failure, optimal taxation, public finance, public policy, required minimum distributions, retirement, taxation. |
