Working Papers

In 1994, Michigan enacted a comprehensive school finance reform that not only significantly increased state aid to low-spending districts, but also placed significant limits on local discretion over school spending. These limits especially constrained the high spending districts. This scenario affords us a unique opportunity to study the implications of such reforms on resource allocation, particularly as they differentially affected districts situated at different points of the pre-reform spending distribution. We find that the reform generally led to a negative effect on the growth of instructional expenditure and its share, as well as in teachers per pupil. But these declines were sharpest in the high spending districts. Interestingly, while trends for shares of administration expenditure as well as administrators per pupil also showed across the board declines, these declines were actually the smallest for the high spending districts. To the extent that instructional expenditures are more productive and contribute to student achievement more than administrative expenditures, these results suggest that loss of discretion acted as a disincentive for districts located throughout the spending distribution. Moreover, this disincentive effect was the strongest in the high spending districts. These findings have important policy implications and suggest that school finance reforms (or other policies) that place significant restraints on local discretion can lead to unintended disincentive effects, which should be taken into account while devising policy.

As policymakers strive to improve student achievement, school turnaround and portfolio management reforms are growing in popularity. The Los Angeles Unified School District Public School Choice Initiative (PSCI) combines these reforms. Using student-level difference-in-difference regressions alongside qualitative analyses, this paper examines the effectiveness of PSCI in improving student achievement in the first year of its implementation. We find that students in PSCI "focus" (turnaround) schools perform significantly worse than their peers in comparison schools on math and English Language Arts achievement tests. However, students in PSCI "relief" schools that are newly opened to relieve overcrowding in surrounding schools perform significantly better. Our qualitative analyses suggest that there are many factors associated with the implementation and execution of the intervention that may contribute to these results.

Using the propensity score matching method and regression models with data from the Education Longitudinal Study of 2002, this study found a significant Catholic school effect on the mathematics achievement of 12th graders who were least likely to attend a Catholic school. These students tended to come from low socioeconomic backgrounds. A significant Catholic school effect was also found for students studying in districts that allowed private education through public funding, regardless of their likelihood to attend a Catholic school. Among students least likely to attend Catholic school, only those who were eligible for public aid for private schooling received academic benefits from Catholic schooling. The results from school districts have implications for policies concerning Catholic schools.

This paper examines the location of charter schools in New York State. We begin by identifying a set of location incentives created by charter school financing and accountability provisions, some of which are unique to New York and others of which are inherent to charter schools. Estimated Poisson and Tobit regression models reveal that the pattern of charter school locations across districts are highly consistent with incentives created by financing and accountability policies. Particularly, we find that charter schools are significantly more likely to locate in districts with high operating expenses per pupil, and thus, high charter school payments; low teacher costs; and low performance. Charter schools are also more likely to locate in districts with concentrations of college educated adults as well as high levels of diversity in educational attainment. Within districts, charter schools tend to locate near areas with concentrations of low-income and minority students, who otherwise might have constrained educational choices, which suggests that concerns about the costs of enabling low-income students to reach achievement standards do not discourage charter schools from locating near concentrations of disadvantaged students. This analysis will be of interest to state legislators and their staff, state education department officials and charter school authorizers.

This article examines several major features of local school markets in two mid-sized municipalities in Sweden. Empirical material collected for the study consists of interviews with 81 school stakeholders and an extensive review of official documents, statistics and literature. Both parents and teachers interviewed in the study have reported several distinctly positive outcomes of school competition, such as pedagogical improvements, new profiles, increased communication and better cooperation between teachers and parents. However, they also described increased segregation, stress attached to making a choice, excessive time devoted to marketing and grade inflation as a way to improve a school’s position in the market. I argue that a) even if independent schools appear to be winners, the organizational habitat of a school is not exclusively correlated with its position in the market and that b) a school’s ranking is critically dependent on its neighborhood, the representation of highly educated parents, the number of immigrant students enrolled and the reputation and status of the school in the community. The overall conclusion is that the practical implementation of school choice policy in these two municipalities has both built on and continued to promote the socially unjust and segregating foundations of the educational system.

Charter schools are changing public and private school enrollment patterns across the United States. This study analyzes district-level enrollment patterns for all states with charter schools, isolating how charter schools affect traditional public and private school enrollments after controlling for changes for the socioeconomic, demographic, and economic conditions in each district. 
Private school enrollments are much more sensitive to charters in urban districts than in non-urban districts. Overall, about 8 percent of charter elementary students and 11 percent of middle and high school students are drawn from private schools. In highly urban districts, private schools contribute 32, 23, and 15 percent of charter elementary, middle, and high school enrollments, respectively. Catholic schools seem particularly vulnerable, especially for elementary students in large metropolitan areas. 
The flow of private-school students into charters has important fiscal implications for districts and states. When charters draw students from private schools, demands for tax revenue increase. If governments increase educational spending, tax revenues must be increased or spending in other areas reduced, or else districts may face pressures to reduce educational services. The shift of students from private to public schools represents a significant shift in the financial burdens for education from the private to the public sector. 

-This paper was published in Policy Analysis, #707, Cato Institute, August 28, 2012-

A relatively small state, Utah presents an interesting case to study charter schools given its friendly policy environment for charter schools and its significant growth in both the number and the enrollment of charter schools. Although the population in Utah is increasingly diverse, charter school enrollment reflects a significantly lower portion of students of color and students from low-income families. Similar to many states, questions regarding the effectiveness and viability of charter schools continue to be a point of contention. Based on longitudinal student-level data from 2004 to 2009, this paper utilizes two alternative methodologies to evaluate the Utah charter school effectiveness: (a) hierarchical linear growth models with matched sample, and (b) student-fixed effects regressions. Both methods yield consistent results that charter schools on average perform slightly worse as compared to traditional public schools, a result that is primarily affected by the low effectiveness and high student mobility of newly-established charter schools. Interestingly, when charter schools gain more experience they become as effective as traditional public schools, and in some cases more effective than traditional public schools. Given the measured though continuous efforts to expand charter school options, this research has implications for local and state charter school policies, particularly policies that avoid “start-up” costs associated with new charter schools. .

-This is preliminary version of a paper that was published in Economics of Education Review, Vol. 31, No. 5, 835–849 (2012)-

It is generally believed that access to financial aid will increase the likelihood that students will attend and graduate from college. There is a surprising lack of research, however, on the consequences when postsecondary institutions lose eligibility to disburse financial aid. This paper provides what I believe to be the first causal estimates of institution-level financial aid funding loss on enrollment and composition of student bodies. I implement a dynamic regression discontinuity design using a multi-year rule that restricts institutions’ eligibility to offer aid such as Pell Grants and subsidized loans when alumni’s loan repayment rates are below allowed thresholds. Results suggest that financial aid loss discourages enrollment, particularly at for-profit institutions. The decline in enrollment appears to be driven by fewer new enrollees. I find less conclusive evidence that ineligibility to disburse federal financial aid substantially alters student body composition. This research is particularly relevant considering recent federal rulemaking that will further limit the number of institutions eligible to disburse financial aid based on additional student loan debt repayment requirements. Restrictions such as these are intended to protect students and the integrity of federal aid programs, but may also have implications for access to higher education.ly and more broadly, with robust evidence in favor of improvements in all AYP objectives.

NCLB mandated the institution of AYP objectives, and schools are assigned an AYP pass/fail based on performance in these objectives. AYP-fail status is associated with negative publicity and often sanctions. In this paper, I study the incentives and responses of schools that failed AYP once. Using data from Wisconsin and regression discontinuity designs, I find evidence of improvements in high stakes reading and spillover effects to low stakes language arts in these schools. The patterns are consistent with focus on marginal students around the high stakes cutoff, but this did not come at the expense of the ends. There is not much evidence in favor of improvement in high stakes math, or low stakes science and social studies. Performance in low stakes grades suffered, and so did performance in weaker subgroups inspite of their inclusion in AYP computations. While there is no robust evidence in favor of effects in test participation and graduation, attendance improved in threatened schools where they mattered for AYP. There is strong evidence in favor of response to incentives—schools that failed AYP by failing only in reading and/or math did substantially better in these subject areas. Credibility of threat mattered—AYP-failed schools that faced more competition responded both more strongly and more broadly, with robust evidence in favor of improvements in all AYP objectives.

The primary education sector in Pakistan faces many challenges relating to access to education and quality of resources. This paper evaluates the Educational Voucher Scheme (EVS) in Lahore, Pakistan aimed at increasing access to primary schooling for low income families residing in the underdeveloped areas of Lahore by using the four criteria for evaluating privatization plans in education outlined by Levin (2002): Freedom of choice, equity, productive efficiency and social cohesion. The study finds that the EVS is associated with greater choice for families as they move from a situation of little or no schooling options to a situation of many schooling options. Similarly, the EVS increases equity by providing lower income families with access to private schools, and because, as studies of private schools in Pakistan have found, that on average, private schools are associated with better resources and academic outcomes particularly for girls. Student achievement data from the EVS is limited however, the sample shows that on the whole EVS students are doing no worse than their non-EVS peers. Furthermore, studies of private schools in Pakistan show they may have a cost advantage over public schools due to lower teacher wages in the private sector therefore, the EVS could potentially lead to increased productive efficiency in the form of higher student achievement and lower school costs by giving families the opportunity to send their children to private schools. Lastly, due to the lack of regulations on participating schools, social cohesion may not be achieved under the EVS as there is no way to ensure that students experience a common educational experience. However, if individuals in society are expected to possess some minimum level of literacy to participate in social and economic institutions, then it can be argued that the EVS could increase social cohesion by equipping children with basic literacy skills.

Proponents of charter schools claim that a highly competitive school choice environment will increase student performance. Results in the existing literature are mixed, without a clear pattern across states on the impact of charter schools on traditional public schools (TPS). However, all previous studies have ignored the heterogeneity in the types of charter schools that compete with TPS for students. Using longitudinal, student-level data from Milwaukee public schools (MPS), I estimate the competitive effects of charter schools sponsored by different authorizers on the outcomes of students attending TPS. Identification comes from the longitudinal variation in competition levels generated by the entry and exit of these different types of charter schools. I find that non-district sponsored charter schools have significant positive effect on students’ math and reading achievement in neighboring public schools. However, only in math, this non-district sponsored charter school competitive effect is statistically different from the competitive effect of district sponsored charter schools. Secondly, this competitive effect varies across subgroups of students. Further, additional research is necessary to assess whether these findings can be attributed to competition from high quality charter schools or other possible consequences of higher autonomy from the school district. At least in Milwaukee, I can conclude that a competitive school market with non-district sponsored charter schools is beneficial to some subgroup of students without hurting other subgroups.

There is widespread concern that administration consumes too much of the educational dollar in traditional public schools, diverting needed resources from classroom instruction and hampering efforts to improve student outcomes. By contrast, charter schools are predicted to have leaner administration and allocate resources more intensively to instruction. This study analyzes resource allocation in charter and district schools in Michigan, where charter and tradition public schools receive approximately the same operational funding. Holding constant other determinants of school resource allocation, we find that compared to traditional public schools, charter schools on average spend nearly $800 more per pupil per year on administration and $1100 less on instruction.