School Qualities, Education Policy, and the Distribution of Educational Benefits

Dennis Epple和Richard E. Romano卡内基梅隆大学SBR-9601207本研究将从理论角度研究公立学校政策。开发了一个与理论模型并行的计算模型，并用于探索模型的性质和研究政策预测。我们比较了有居住要求的地区（社区学校）和在公立学校中有一定程度选择的地区的结果。选择公立学校无成本的结果将与选择公立学校有成本的结果（例如通勤成本）进行对比。在实践中，当学校系统引入选择课程时，他们通常会对不同学校的入学标准施加选择标准；该研究将调查这种“受控选择”政策如何影响均衡结果。这些社区学校和选择方案将在学校财政集中和财政分散的环境中加以考虑。我们分析了只有公立学校的情况，然后分析了公立学校和私立学校都有选择的情况。该研究描述了在所有不同模式下决定所有学校支出的因素。公立学校的开支将由多数决定原则来决定，利用我们之前在有私人选择的情况下对公共提供进行投票的工作。本研究开发的模型具有以下特点：学生的特点是能力水平和家庭收入。学生的教育成就在自身能力、每个学生的教育支出以及学生同学的能力方面都在增加（同侪群体效应）。在学生的家庭收入中，对教育成就的需求正在增加。该项目的最后一部分将增强学校质量的特征，包括教师与学生的比例和教师技能的层次结构。将扩大教育成绩功能，允许平均教师素质和师生比影响成绩。教师的效用函数将取决于收入、学生质量、教师同行的质量，以及（可能的）教师/学生比例。这个模型也可以用来分析学院和大学教育。我们将调查不同的政策如何影响不同能力和收入水平的学生的成绩，学生在学校之间的能力和收入分布，学生的福利，学校的税收和支出，私立学校的财政援助政策，教师在学校之间的能力分布，以及所有学校教师工资的确定。

Dennis Epple and Richard E. Romano Carnegie Mellon University SBR-9601207 This research will study public school policy from a theoretical perspective. A computational model paralleling the theoretical model is developed and used to explore properties of the model and investigate policy predictions. We compare outcomes in localities that have residency requirements (neighborhood schools) with outcomes where there is some degree of choice among public schools. Outcomes with costless choice among public schools will be contrasted to outcomes with costly exercise of public school choice (e.g. commuting costs). In practice, when school systems introducing choice programs, they often impose selection criteria for admission to various schools; the research will investigate how such "controlled choice" policies affect equilibrium outcomes. These neighborhood school and choice alternatives will be considered in environments with centralized school finance and with decentralized finance. We analyze settings with only public schools, then when there are both public and private school alternatives. The research characterizes the factors that determinate expenditures in all schools in all variations of the model. Public school expenditures will be determined by majority rule, exploiting our earlier work on voting over public provision when there are private alternatives. The model developed in this research has the following features. Students are characterized by an ability level and household income. A student's educational achievement is increasing in own ability, educational expenditure per student, and in the abilities of the student's classmates (a peer-group effect). The demand for educational achievement is increasing in the student's household income. The final part of the project will enhance the characterization of school quality to include faculty-student ratios and a hierarchy of faculty skills. The educational achievement function will be broadened to allow average faculty qu ality and faculty-student ratios to affect achievement. The utility function of faculty will depend on income, student quality, quality of faculty peers, and, possibility, the faculty/student ratio. This model will serve to permit analysis also of college and university education. We will investigate how alternative policies affect the achievement of students of varying ability and income levels, the distribution of students by ability and income across schools, the welfare of students, school taxes and expenditures, private school financial aid policies, the distribution of faculty by ability across schools, and determination of faculty salaries in all schools.