Distinguishing the Roles of Poverty and Homelessness in Long-Term Educational and Early Labor Market Outcomes in Maryland

Background and Purpose: Poverty is a common cause of homelessness, an extreme form of residential instability, affecting more than 1.5 million U.S. public school students between the 2015-16 and 2017-2018 school years (Fantuzzo et al., 2012; National Center for Homeless Education, 2020). Although homeless and low-income students share similar experiences (Buckner, 2008), the “continuum of risk” framework suggests that homeless students are more likely to experience worse outcomes due to increased exposure to multiple co-occurring risks linked to poverty (Brumley et al., 2015). Yet, few studies have managed to disentangle their distinct effects on long-term academic and workforce outcomes in relation to the continuum of risk (Buckner, 2008). Within the homeless student population, adolescents are considered to be the most “at-risk” subgroup (Toro et al., 2007); however, the vast majority of research has focused on children, resulting in limited knowledge about their unique homelessness experiences (Buckner, 2008). Therefore, study sought to examine the role of adolescent homelessness above and beyond the role of poverty alone and its effects on long-term academic (dropout and college enrollment) and workforce (first-year wages after graduation) outcomes.

Methods: Data were from the Maryland Longitudinal Data System (MLDS), Maryland’s statewide repository for individual-level education and workforce data that are longitudinally linked across three state agencies. Our cohort was comprised of Maryland public school students enrolled in 6^th grade (N = 52,610) in 2007-08 (the earliest year of data available in the MLDS). Students who experienced homelessness at any point between 6^th and 12^th grade constituted about 4% of our cohort (n=2,065) and 44% of students experienced poverty (e.g., eligibility for free or reduced-price meals) without ever being identified as experiencing homelessness by their school system. This study used Stata/SE version 15 to obtain descriptive statistics and models were fitted using Markov Chain Monte Carlo procedures in MLwiN version 3.02.

Results: Overall, homeless students experienced worse academic outcomes than their low-income, stably housed peers, but did not experience worse early labor market outcomes. Compared to their low-income, stably housed peers, homeless students were twice as likely to drop out of high school and 31% less likely to attend college. Within the first year after on-time, high school graduation, no significant differences were observed in workforce earnings between homeless youth and their low-income, stably housed peers for both non-college and college enrollees. However, meaningful differences were observed when examining earnings by race, with White homeless students earning significantly lower wages than their stably housed, low-income peers. On the contrary, Black or Other racial/ethnic homeless youth and their stably, housed, low-income peers who were not enrolled in college earned similar wages, despite poverty and/or homelessness experiences, in which their wages were uniformly lower than their White counterparts.

Conclusions and Implications: Findings suggest that homeless and low-income, stably housed students have distinct risks affecting academic outcomes but overlapping risks for early labor market wages for both non-college and college enrollees. Additionally, findings highlight significant racial disparities in early workforce wages for homeless and low-income, stably housed minority youth.