Johnson, Rucker C.
Long-Run Impacts of School Desegregation & School Quality on Adult Attainments NBER #16664
University of California - Berkeley
What are the long-run impacts of court-ordered school desegregation plans on adult attainments?
Journal Name or Institutional Affiliation:
NBER -National Bureau of Economics
Unpublished institutional report (e.g., NBER, HCRP, Brookings)
- For education attainment: each additional year of exposure to court-ordered desegregation leads to a 1.3 percentage point increase in likelihood of graduation from high school for blacks. Each additional year of exposure to court-ordered desegregation leads to a 0.08 increase in years of education for blacks. There are no significant effects for whites.
- Men’s Labor Market Outcomes & Adult Family Income and Poverty Status: an additional year of exposure to court-ordered desegregation significantly increases black men’s annual earnings by roughly 5 percent, which is a combination of a 2.9 percent significant increase in wages and an annual increase in work hours of 39 hours. Among black men and women, an additional year of exposure increases the family income-to-needs ratio by about 0.1 and reduces the annual incidence of poverty in adulthood by 1.6-1.9 percentage points.
- Probability of Incarceration: for black students, relative to growing up in segregated schools, exposure to desegregation beginning in elementary school leads to a 22.5 percentage point reduction in the probability of deviant behavior , a 14.7 percentage point reduction in the probability of incarceration by age 30 and 3.8 percentage point decline in the annual incidence of incarceration during ages 20-34.
- Adult Health Status: an additional year of exposure to court-ordered desegregation increases adult health status index for blacks between 0.3-0.6 points.
Academic Achievement, Desegregation, Earnings, Long Term Outcomes, Occupational Outcomes, School Quality
Method of Analysis:
Difference in Difference Regression, Fixed Effects Regression Models, Two Stage Least Squares
Panel Study of Income Dynamics (PSID) sample members born between 1950 and 1975 as well as the schools they attended
Unit of Analysis:
Quantitative-Longitudinal, Quantitative-Panel Data
For data on students: 4,683 children were followed into adulthood of the original 7,212 children in the PSID survey sample. These individuals were matched with their childhood residential location for the school measures.
- For data on school resources: 1,073 different neighborhoods from 186 school districts in 33 different states were identified and characterized by census block data.
- Dependent variables includes: educational attainment, adult earnings, wages, annual work hours, family income and poverty status, whether ever incarcerated, and adult health status.
- Independent variables are school desegregation and school quality.