Association for Public Policy Analysis & Management

Every year, 13 percent of all public school students receive special education services (U.S. Department of Education, 2015). These services provide special needs students with an “appropriate education that meet[s] their unique needs" aimed at facilitating short-run academic success and preparing students ``for further education, employment, and independent living" (Individuals with Disabilities Improvement Education Act (IDEA) of 2004). Much research and policy attention has focused on whether special education services are being appropriately assigned. For instance, the over-representation of minority groups in special education programs is regarded as the “most long-standing and intransigent issues" in the field of special education (Skiba et. al, 2008). However, very little is known about how this placement (or lack of placement) impacts special needs students, particularly during adulthood.

This research builds upon previous work focused on the short and medium run impacts of special education placement (Setren, 2016; Cohen, 2007; Hanushek, Kain, & Rivkin, 2002; Prenovitz, 2017), by providing the first causal estimates on the long-run impacts. This is achieved by exploiting a 2005 policy change in Texas that introduced a district enrollment target of 8.5 percent into special education programs. This threshold was far below the percentage of students being served in these programs at the time, when the average district served 13 percent and a quarter of all districts served more than 16 percent of their students in such programs. In order to be in compliance, school districts sharply reduced special education enrollment, with the largest reductions for districts furthest from this threshold. This paper utilizes this exogenous variation in access (measured by a students' district's distance from 8.5 and years since 2005) and longitudinal administrative data for the population of Texas public school students to causally identify the impact of special education placement during K-12 on adult outcomes.

Policy-exposed special education students are 4% more likely to experience service removal at the average school district. However, there are significant differences across different types of students. Students with more malleable diagnoses (who were in more inclusive settings and who were being tested at grade level) and economically disadvantaged students were most likely to experience removal. This removal leads to negative long run outcomes. Policy-exposed students are 2% less likely to graduate from high school and 4% less likely to enroll in college. Moreover, these effects are larger for those students most likely to experience service removal. These findings suggest that students on the margin of special education placement benefit from greater access to such services.

Federal laws and state-level policies are often responsible for significant changes in special education access (Cohen, 2007; Cullen, 2003; Cullen & Reback, 2006; Figlio & Getzler, 2006; Jacob, 2005). Despite this, there is little evidence on how such changes in access impact student outcomes. By focusing on this Texas policy, this paper contributes to our understanding of how students on the margin are impacted by changes in access to these programs.