Title: The influence of private school enrollment on public school performance
Abstract: Abstract School choice reform refers to changes that allow parents greater freedom to choose schools for their children. School choice reform is contentious because it fundamentally alters the environment in which public and private schools operate and could result in significant changes for both demanders and suppliers of education. This article develops a model of public education with imperfect exit to predict how private school enrollment influences performance of public schools. Empirical evidence from data on all private and public schools in California provides substantial support for the hypothesis that public school test scores are inversely related to private school enrollments and charter school enrollments when private and charter schooling reflects exiting by parents unhappy with local public schools. Implications regarding how expanded private school choice might influence public school performance in California and elsewhere are discussed. Acknowledgements I am indebted to an anonymous referee for many excellent comments. Notes 1Martinez-Vazquez and Seaman (Citation1985) and Hamilton and Macauley (Citation1991) find that, when communities with diverse populations have relatively few school districts, there are more private schools. Evidence then suggests that private enrollments are higher when public schools do not support diverse preferences and so greater numbers of private schools provide more exit opportunities for demanders. 2Assumptions include perfect information and mobility and absence of free riders and interjursidictional externalities. 3See West (Citation1990) for a discussion. There may also be some ability to petition to attend another public school. 4See National Center for Education Statistics (2002). 5See, for example, Alt and Peter (Citation2003). However, it should be noted, that Coleman et al. (Citation1966), or the so-called Coleman Report, concluded that the strongest predictor of student performance was family background as, for example, measured by household income and parental socio-economic status. Therefore, conventional measures such as per pupil spending and student-to-teacher ratios were found to be relatively unimportant in most cases. This report prompted numerous studies that mostly supported the original findings of the Coleman Report as discussed in Hanushek's (1986) extensive survey of the literature. More recently, others have argued that school effects such as per pupil funding are more important than previously believed. See, for example, Card and Krueger (Citation1992) and Hedges et al. (Citation1994). Several researchers doubt the validity of recent studies concluding that school effects are important; see, for example, Betts (Citation1995), Grogger (Citation1996) and Hanushek et al. (Citation1996). While debate continues on how important school effects are, most researchers appear to agree that nonschool effects such as family and community background are quite important when it comes to determining student performance. Therefore, perhaps school effects are less important as well than nonschool effects in explaining student performance differences between private and public schools." 6Quote taken off http://www.nea.org/vouchers/index.html during December 2005. 7The following quote from NEA's website http://www.nea.org/vouchers/index.html during December 2005 makes it clear that they view vouchers as a threat to poorly-performing public schools: 'NEA and its affiliates support direct efforts to improve public schools. There is no need to set up new threats to schools for not performing. What is needed is help for the students, teachers, and schools who are struggling.' 8Data from Statistical Abstract of the United States, 2003. 9In 1999–2000, approximately 27000 private schools, with 404000 full-time-equivalent (FTE) teachers, enrolled 5.3 million students. These schools accounted for 24% of all schools in the United States, 10% of all students, and 12% of all FTE teachers; see US Department of Education (Citation2002). 10Numbers discussed here are from California Department of Education (2004). 11Counties having 10% or more of their total students enrolled in private schools in 2003−2004 were: Alameda (11.8%), Contra Costa (10.3%), Los Angeles (10.5%), Marin (18.7%), Napa (13.4%), Orange (10.1%), San Francisco (29.1%), San Mateo (15.4%), Santa Clara (12.4%) and Santa Cruz (11.2%). 12These defeats are consistent with Kenny's (2005) discussion of initiatives that concludes that voucher bills have been passed only in the more conservative Republican states. 13Alpine County had a total public school enrollment of only 148 in 2003. 14For instance, see the review of more than 35 studies assessing public school performance in Belfield and Levin (Citation2001). Also, Lankford and Wyckoff (Citation1992) find that parents are sensitive to school quality as measured by test scores of public and private elementary schools. 15Educational attainment is measured by the percentage of adults over the age of 25 with at least a bachelor's degree in 2000 and is published in the Statistical Abstract of the United States, 2003. Results did not change when educational attainment was substituted for income. 16See, for example, Hamilton and Macauley (Citation1991) and Schmidt (1992). 17See, for example, Peng and Hill (Citation1995) and Demie (Citation2001). 18A weak market share for private schools may result in competitive pressures that do not reach critical thresholds necessary for influencing behaviour of public schools. Hoxby (Citation2003) discusses this possibility. Vouchers and tax-preferences that lower costs of private enrollments would lower price differentials between private and public schooling and likely lead to a greater competitive presence from private schools. Marlow (Citation1997, Citation2000), Dee (Citation1998), Hoxby (Citation2002) and Greene and Kang (Citation2004) report evidence in support of the hypothesis that various forms of competition improve public school performance. Although opposing evidence is reported in Sander (Citation1999) and Jepsen (Citation2002), Belfield and Levin's (2001) summary of the literature concludes that a majority of studies find that competition improves public school performance.