Swedish School Results, Student Background, Competition and Efficiency

Christophe Andr - MSc, Senior Economist, OECD Economics Department.E-mail: christophe.andre@oecd.orgJon Pareliussen - MSc, Economist, OECD Economics Department.E-mail: jon.pareliussen@oecd.orgHyunjeong Hwang - PhD, Statistician, OECD Economics Department.E-mail: hyunjeong.hwang@oecd.orgAddress: OECD, Economics Department, 2 rue Andr-Pascal, 75775 Paris Cedex 16, France.Sweden's declining results in the Programme for International Student Assessment (PISA) for 15-year olds and other international tests between 2000 and 2012 have raised concern about the efficiency of the Swedish school system, even though results improved recently. Furthermore, inequality in educational outcomes between socio-economic groups have widened. A specificity of the Swedish school system is that it allows free choice between public and private schools. This has triggered a lively debate on the implications of competition for school results and educational inequality. Against this backdrop, this paper presents an econometric analysis of lower secondary school performance in Sweden, using a panel covering most schools in the country over the period 2013-17. We find that for-profit private schools underperform non-profit and public schools on average, although with large heterogeneity. School competition is associated with lower results in schools with a high share of pupils from weaker socio-economic backgrounds, which is consistent with negative peer effects in left-behind schools. Panel Stochastic Frontier Analysis points to a relatively narrow distribution of inefficiency across schools, with relatively few schools performing very poorly after controlling for their resources and the socio-economic background of their pupils. These results call for better targeting resources towards supporting the pupils most in need and steering competition and school choice so that they benefit pupils from all socio-economic groups equally.