Essays on taxation: Positive and normative aspects

This dissertation consists of three essays on taxation. The first essay examines whether government policy, through the use of tax incentives, is able to encourage household savings. This essay analyzes the impact of the 1990 Education Savings Bond Program. This policy created an additional tax incentive for owners of existing government savings bonds, by allowing interest earnings to be exempt from income taxes in years where the household incurs a qualified education expense. Using the 1989 and 1992 Survey of Consumer Finance data sets, a difference-in-difference methodology is used to measure how household savings has changed over time for households with college-bound children as opposed to those without. Households without college-bound children do not need to save for education and thus are not affected by the program. The comparison of savings for the two groups, correcting for individual characteristics, reveals the impact of the Education Savings Bond Program. In addition, this procedure allows one to infer if there has been a crowding out effect due to the Education Savings Bond Program. The results indicate that the policy has not had an effect on household savings.;The second essay uses two different estimation procedures to calculate the incidence of environmental taxes and compares the results. Both estimation procedures assume non-separability of leisure and so the labor response is incorporated into estimates of household behavior. The first method is the Almost Ideal Demand System (AIDS) model of Deaton and Muellbauer. The AIDS model assumes linear Engel curves and if this assumption is violated then welfare estimates are biased. The Quadratic Almost Ideal Demand System (QUAIDS) model of Banks, Blundell and Lewbel extends the AIDS model to allow for non-linear Engel curves. Households consume three goods---a composite clean good, a composite energy good and leisure. Data on household consumption is from the Interview Survey component of the Consumer Expenditure Survey. The AIDS model finds the energy good and leisure to be substitutes while the QUAIDS model finds no relationship between the two goods. Moreover the AIDS model is found to overestimate the welfare loss of environmental taxes on low-income households but underestimate the welfare loss of environmental taxes on high-income households.;The third essay again uses the Almost Ideal Demand System model of Deaton and Muellbauer to estimate demand for junk food as well as calculate the incidence of taxes on junk food. The model assumes households consume three goods---a composite healthy food good, a composite unhealthy food good (junk food) and a composite nonfood good. Data on household consumption is from the Diary Survey component of the Consumer Expenditure Survey. The Diary Survey collects detailed data on food expenditures. Price data consists of price indices for various commodities which is available from the Bureau of Labor Statistics. The compensated own-price elasticities indicate that both healthy food and junk food have inelastic demand. In addition, healthy food and junk food are found to be substitutes. The elasticity values found are consistent with the literature.