Paying for Progress: Conditional Grants and the Desegregation of Southern Schools

Abstract

This paper examines how a large conditional grants program influenced school desegregation in the American South. Exploiting newly collected archival data and quasi-experimental variation in potential per-pupil federal grants, we show that school districts with more at risk in 1966 were more likely to desegregate just enough to receive their funds. Although the program did not raise the exposure of blacks to whites like later court orders, districts with larger grants at risk in 1966 were less likely to be under court order through 1970, suggesting that tying federal funds to nondiscrimination reduced the burden of desegregation on federal courts.

Decomposing School Resegregation

“…As a result, many districts involved in these decisions returned to neighborhoodbased school assignment policies, leading to increases in racial imbalance and racial isolation (An and Gamoran 2009;Clotfelter, Vigdor, and Ladd 2006;Mickelson 2005;Reardon et al 2012). Also to blame are federal policies, which helped spur desegregation in the 1960s by tying federal funds to compliance with nondiscrimination laws (Cascio et al 2010), but have increasingly favored accountability policies and turned a blind eye to school segregation (Orfield and Lee 2007). A third factor is the recent expansion of school choice (Orfield et al 2002); evidence suggests that families choose charter, magnet, and other choice schools in ways that exacerbate racial imbalance and isolation (Bifulco and Ladd 2007;Saporito and Sohoni 2006;Sikkink and Emerson 2008).…”

School Segregation, Educational Attainment, and Crime: Evidence from the End of Busing in Charlotte-Mecklenburg *

“…The seven-year change in the exposure index (a commonly used measure of segregation) was about 13 percentage points, which is nearly as large as the effect of court-ordered desegregation in the 1960s and 1970s (Rossell and Armor 1996;Guryan 2004). Second, studentlevel data from a single district coupled with a quasi-experimental research design allow us to explore how changes in school racial composition might affect students through different mechanisms such as school resources and peer effects (Reber 2010;Cascio et al 2010;Johnson 2011;Bifulco et al 2011;Imberman et al 2012). Third, we add to the literature on the effect of school and neighborhood interventions on long-run outcomes such as college attendance and crime (Katz et al 2001;Ludwig et al 2005;Kling et al 2007;Angrist et al 2011;Deming 2011;Chetty et al 2011;Deming et al 2013;Dobbie and Fryer 2013;).…”

From Brown to busing

“…54 The same patterns of more court involvement and less intensive deseg-53 However, for neither COURTORDER nor FRACALLBLACK can we reject the null hypothesis that the coefficients on FRACPOOR are the same in 1964 and 1966. In other work, we use a more idiosyncratic component of the variation in Title I grants across districts to estimate the causal effect of financial incentives on desegregation directly (Cascio et al, 2008). 54 When compared to districts in the first decile of black share (less than or equal to 9.5 percent), districts with black shares between 16.0 and 34.2 percent (deciles three to five) and over 46.6 percent (deciles eight to ten) expe- regation among blacker districts persisted through 1968 (Fig.…”