Are Charter Schools Draining Private School Enrollment?

Ω

Rajashri Chakrabarti, Joydeep Roy, and Elizabeth Setren*

Charter schools are a major policy initiative at the national and local levels. As charter schools spread, one key question is whether they reduce private school enrollment, especially at Catholic schools. If so, an increase in charters could change public school spending patterns, decrease the number or size of private schools, and alter educational outcomes and school quality for public and private school students. But is this really the case? Maybe not. In this post, based on our 2010 New York Fed staff report, we find that despite widespread fears to the contrary, the expansion of charter schools in Michigan led to only a small decline in private school enrollment.

    Charter schools are publicly funded schools that are exempt from many of the rules, regulations, and statutes that apply to other public schools in exchange for some accountability requirements that are set forth in each school’s charter. Since the first charter school opened in Minnesota in 1991, the schools have expanded rapidly. There are now more than 5,000 charter schools throughout forty U.S. states and the District of Columbia. From 1999 to 2010, enrollment in charters across the nation more than quadrupled and now totals more than 1.5 million students.

    There is a general impression that charter schools have negatively affected private school enrollment, especially in Catholic schools. Our staff report analyzes whether the introduction of charters in Michigan drained students from local private schools. In December 1993, Michigan adopted one of the strongest charter school laws in the nation, and the first charter school in the state opened in the 1995-96 school year. We use Michigan private school data from the 1989-90 to 2001-02 school years and Michigan charter school data from the 1995-96 to 2001-02 school years. Enrollment information from before the charter law was adopted is used to control for preexisting trends in private school enrollment. And we use data from the years after the charter schools opened to determine if their impact increases or decreases as they age. Our focus is on elementary schools because charter schools mostly cater to elementary grades.

    We find that while charter schools led to a fall in private school enrollment, the decline was modest. We also find that with the passage of time, as the charter sector matures its effect on private schools increases. Having a charter school within a two-mile radius decreases private school enrollment by about 1.2 percent per year. An increase in charter enrollment by one student within the same radius decreases private school enrollment by 0.01 percent per year. A private elementary school in Michigan had an average enrollment of 156 students during the period. Furthermore, there is no evidence that Catholic schools lost more students to charter schools than did secular private schools.

    Thus, our paper’s findings suggest that charter schools did not lead to a major crowding out of private school enrollment in Michigan. Instead, charters led to only a small decline in private school enrollment. These findings are very robust and survive a variety of sensitivity tests.

    It is interesting to consider why charters do not drain private school enrollment, even though many observers assert that they do. Several factors could explain this outcome. The modest impact could be a function of charter school quality. In a study of Michigan charter schools, Eric Bettinger finds that test scores of public school students who moved to charter schools did not improve compared with the scores of their public school counterparts. If parents did not perceive charter schools as a better option than public schools, they would not move their kids from private to charter schools. If the perceived quality of Michigan charters, relative to public schools, is lower than that of other states, there might be more evidence of students transferring from private to charter schools outside of Michigan.

    Additionally, unlike in our study, enrollment levels in charter and private schools, often reported in the popular press, do not reflect causal trends. These numbers might be impacted by preexisting trends in private enrollment, regular mobility of students between schools, and state- or district-wide policies or events. Furthermore, the location of charter schools is not random. Charter schools are likely to be established in towns where parents are dissatisfied with the private and public school options. These factors are not accounted for when we simply look at the raw enrollment numbers. Rather, we need a causal analysis that accounts for these factors, as was done in our study.

    Similarly, the decline in Catholic parochial school enrollment in areas near charter schools often reported in the media does not necessarily imply that charters caused the change. The decline could be part of a larger trend and not due solely to the presence of charter schools. It could also be the result of recent tuition increases in Catholic schools or other changes relating to Catholic schools.

    In summary, our staff report shows that charter schools in Michigan have led to only a modest decline in private school enrollment in the state. Moreover, they have not affected Catholic school enrollment more adversely. These findings may dispel some concerns about the impact of charter schools on public school spending and the health of local private, especially Catholic, schools.


*Joydeep Roy is a senior economist at the Independent Budget Office and a visiting professor of economics and education at Columbia University; Elizabeth Setren is a research associate in the Federal Reserve Bank of New York’s Research and Statistics Group.