New evidence from Norway
Hans Bonesrønninga, , Linn Renée Naperb
aNorwegian University of Science and Technology, Norway
bCentre for Economic Research at NTNU, Norway
Date of submission: February 12., 2008
Abstract: We examine whether potential competition effects from private schools differ across student subgroups in Norwegian public schools. Competition is measured by the share of students attending private schools. This measure is potentially endogenous to student achievement, and is treated by IV estimation. Sorting of parents across municipalities is considered by exploiting the location of universities and colleges. Indications of non-uniform competition effects are provided: while there is no evidence that students with less educated parents gain from competition, the effect for students with highly educated parents is positive. The exact size and significance level of the effect for the latter subgroup is hard to pin down.
Keywords: school competition; private schools; student achievement JEL-classifications: I21, H42
There are two major ways charter schools might improve the average student achievement in an economy: by being better than public schools; and by improving the performance in public schools by introducing more school competition. The present paper evaluates the second effect. It is a controversial issue whether school competition will work as intended. From a theoretical point of view, the nature of competition between schools is not clear-cut: The education market seems to be a differentiated product market, where schools differ in their characteristics and where parental preferences over school characteristics differ. One conjecture is therefore that the degree and character of competition will vary with the characteristics of both suppliers and demanders. Not surprisingly then, the existing empirical literature on school competition – between public and private schools or between public schools – does not provide any sweeping and clear-cut conclusions (see for instance, Hoxby 1994, 2000, Dee 1998, Fiske and Ladd 2001, Levačić 2004, Bettinger 2005, Sandstrøm and Bergstrøm, 2005, Bayer and McMillan, 2005, Gibbons, Machin and Silva, 2005, Hsieh and Urquiola, 2006, Card, Dooley and Payne, 2006).
The present paper adds to the existing literature by providing empirical evidence from Norway. First, we investigate whether the overall performance of public schools in Norway is affected by competition from private schools. Thereafter, we ask whether the competition effects differ across student subgroups. Our hypothesis is that principals and teachers within public schools might realize that they can influence parental school choices by reallocating resources and/or attention towards the students they think are most likely to opt out. The incentives to reallocate resources or attention might vary with the attractiveness of the most mobile students (it is not unlikely that mobility and attractiveness is correlated). Thus, if the public sector actors’ responses to competition are to reallocate resources between student subgroups, we expect the attractive and mobile student subgroups to gain, and the less attractive and less mobile students, to lose from competition.
The private school sector in Norway is small, and has some distinct characteristics. Most privately owned schools are religious schools or schools with an alternative pedagogy, reflecting the Private School Law of 1970 which offers only these kinds of private schools public funding. Private schools are offered (almost) free of charge to the students. The municipalities experience a decrease in their budgets when students exit for private schools; implying that the financing of private schools very much portrays a voucher system. Private schools are not allowed to discriminate between applicants, but the rules that can be applied when schools are...