Additional Funding for Special Needs Students: Quasi-experimental Results From the Netherlands

We compare the mental health and academic performance of Dutch special needs students and regular students before and after the introduction of a school inclusion policy in 2003. The school inclusion policy consisted of a personal budget for eligible special needs students to support progression in regular education. We use five waves of data between 2000 and2013 from 1,112 students, aged 10 to 12 at baseline, of the TRacking Adolescents' Individual Lives Survey. We use validated instruments to measure mental health and academic performance. Following norms for eligibility of the Dutch Ministry of Education, Culture and Science, 295 of the 1,112 students were identified as special needs students. We use a Differences-In-Differences (DID) design to account for time trends unrelated to the policy, enabling direct identification of the effect. Individual-level covariates were included in the extended DID model to reduce the within-group variance. Estimates suggest that the school inclusion policy reduced the inequality in mental health and academic performance between special needs and regular students with 26.2 and 65.1 percent, respectively. Moreover, the achievement gap disappeared before adulthood, as special needs students did not differ in educational attainment at labour market entry from their regular peers. Ten-percent quantile DID estimates show that the policy was the most (least) effective at the lower (higher) end of the mental health distribution. Estimates and statistics are robust to informant bias and propensity score reweighting. Hence, this study shows the potential of supporting progression of special needs students in regular education to reduce inequality in health and human capital. Heterogeneity analyses present a cautionary note, as girls and children from an ethnic minority or a lower socio-economic environment benefited less from the policy.