| This dissertation addresses some issues particular to elementary and secondary education. The first chapter, "Spare the Rod? Principals, Agents, and Corporal Punishment in Public Schools", evaluates public school teachers' relationship with parents and students in a principal-agent setting. Classroom activities are organized as team production that ultimately benefits the students. To maximize their benefits, students as a group delegate decision-making authority to the teacher who encourages cooperation among them and discipline misbehaving participants. Teacher's objective is to maximize her own well-being, which is different from students' objective in absence of any residual claim to the final product of the team. Because teacher monitoring is done indirectly through the political market there is no incentive for public school teachers to spend their resources producing "invisible" output attributes that the students may value very highly. For this purpose this paper focuses on the issue of corporal punishment in elementary and secondary public schools as an opportunity to observe how teacher decisions differ from what students would have deemed optimal. Using a panel of data on 50 US States and the District of Columbia, followed from 1990 to 2000, I find that, in the relevant range, corporal punishment has a negative effect on student performance as measured by eight and fourth graders NAEP mathematics test scores. This result is explained by the fact that although corporal punishment raises the cost of misbehavior, it also lowers students' self-esteem. Because of dissimilar valuation of student self-esteem by the teacher and the student, the net effect is an increase in misbehavior and, thus, a reduction of learning despite of the fact that punishment is intended to accomplish the opposite.; The second chapter, "Cost Savings from Pupil Migration to Private Schools", estimate the magnitude of potential savings to public school boards associated with the migration of public school students to independent schools in connection with voucher or tax credit plans. This paper argues that the appropriate measure of any such savings is the marginal cost of a student. Using three year's worth of school level data for South Carolina, separate cost functions for elementary, middle and high schools are estimated. We find that though some costs are fixed, variable costs comprise about 80 percent of school costs. Moreover, variable costs are very sensitive to changes in enrollment. When estimated in first differences on year-to-year changes, the hypothesis that these costs vary in strict proportionality to enrollment cannot be rejected at generally accepted levels of confidence. We find that these potential savings lie in the range of {dollar}4,000 to {dollar}5,000 per student, which is higher than the amount of withdrawn state support in connection with a tax credit plan. Marginal costs are estimated separately for disabled, gifted and regular students as well.; The third chapter, "Current Demand for Private Education in South Carolina", estimates the elasticity of demand for private education in South Carolina controlling for observable student characteristics, student's milieu characteristics and a measure of geographical distance to nearest private school. Policy makers have concerns about the impact of school vouchers and educational tax credits that operate chiefly by lowering the price of private schools to parents. Analysis of the impact of such proposals thus requires a precisely estimated demand function. (Abstract shortened by UMI.)... |
