The Louisiana Scholarship Program
Contrary to Justice Department claims, student transfers improve racial integration
The Louisiana Scholarship Program (LSP), also known as the Student Scholarships for Educational Excellence Program, provides public funds for low-income students in low-performing public schools to enroll in local private schools. The program was initially piloted in New Orleans in 2008; Louisiana governor Bobby Jindal and the state legislature expanded the LSP statewide in 2012, allowing thousands of public school students to transfer out of their residentially assigned schools and into private schools of their choosing. The program has recently come under fire from the U.S. Department of Justice, which has filed a lawsuit alleging the program is impeding federal school-desegregation efforts initiated in the 1970s. Fortunately, we have data on the school choices made by many voucher recipients, which enables us to study the program’s likely effects on the racial makeup of Louisiana schools.
The evidence suggests that use of private school vouchers by low-income students actually has positive effects on racial integration. Among the subset of students for whom data are available, we find that transfers made possible by the school-choice program overwhelmingly improve integration in the public schools that students leave (the sending schools), bringing the racial composition of the schools closer to that of the broader communities in which they are located. In the school districts under federal desegregation orders, which are the focus of the Department of Justice litigation, LSP transfers improve integration in both the sending schools and the private schools that participating students attend (receiving schools). These findings should help mitigate fears that school choice is harming desegregation efforts in Louisiana.
Setting and Data
The decades since the 1954 U.S. Supreme Court case known as Brown v. Board of Education of Topeka, Kansas, have seen significant judicial oversight of school desegregation efforts in states across the country. In Louisiana, a 1975 case, Brumfield v. Dodd, also declared private schools that segregate or discriminate in admissions ineligible for state funding of any kind. Today, the federal government continues to monitor schools to ensure compliance with desegregation plans. Thirty-four Louisiana school districts remain under federal desegregation orders, and the United States remains a party to desegregation cases in 24 of these districts. The legislation that created the LSP acknowledges that the program is subject to these desegregation plans.
For the 2012–13 school year, approximately 10,000 Louisiana students from low-performing public schools applied for LSP vouchers, which the state awards by lottery based on the number of seats available in local private schools. Nearly 5,000 students used LSP vouchers to enroll in private schools. All of these students were from families with incomes less than 250 percent of the federal poverty line, and approximately 90 percent were African American. LSP vouchers covered students’ tuition at 117 private, mostly Catholic, schools.
The data used in our analysis come from five sources. First, we use unique data from the Louisiana Department of Education on LSP voucher users to track school transfers. Data on private school enrollment and racial composition come from the Private School Universe Survey (PSS), a biennial survey conducted by the National Center for Education Statistics (NCES). We rely on school-level data collected in the 2009–10 school year, the most recent year for which data are available. Corresponding data on public schools in the 2009–10 school year are available from the NCES’s Common Core of Data. We rely on five-year population estimates from the U.S. Census Bureau’s American Community Survey to approximate the school-age racial composition of each Louisiana Core Based Statistical Area (CBSA) in 2009–10. Finally, we identify public school districts under federal desegregation orders using information from the Department of Justice lawsuit, which was filed in August 2013.
Our analysis is limited to LSP voucher users who took the Louisiana state assessments in the 2010–11 school year, meaning they had to have reached the third grade by that time. This restriction is required because our data only allow us to identify an LSP participant’s prior school through the state’s assessment database. This reduces our sample from the 4,954 voucher users enrolling in a private school in the first quarter of the 2012–13 school year to 1,549 students. Because our data on school racial composition come from the 2009–10 school year, our analysis assumes that Louisiana public and private schools did not experience substantial demographic shifts between 2010 and 2013.
Any schools that opened after the 2009-10 school year do not appear in our analysis. Students in such schools represent less than 10 percent of the observations in receiving schools and less than 3 percent of the observations in sending schools. School-age population information is not available for 17 rural counties that are not in CBSAs, and therefore our analysis does not include any LSP voucher students in these counties. These students constitute less than 3 percent of voucher users in our sample. Our analysis excludes LSP voucher users who were participating in the New Orleans–based pilot program prior to 2012–13. These students, representing 30 percent of the original sample, often enroll in the same school they attended the previous year, and therefore do not provide information on the impact of the statewide program.
In total, 4,954 eligible LSP applicants in our data used their voucher to enroll in a private school in the first quarter of the 2012–13 school year. Of these students, 3,133 enrolled in private schools in the New Orleans-Metairie-Kenner metro area. When we apply the restrictions outlined above, we are left with a sample of 841 students. Our final analysis focuses on the 98 percent of these students who are African American, Hispanic, or white. The racial composition of our final analysis sample is similar to that of all voucher users. In particular, African American students constitute more than 80 percent of all voucher users and of our analysis sample, white students approximately 10 percent of both samples, and Hispanic students less than 5 percent of either sample.
Our primary analysis seeks to answer the question, What is the impact of transfers directly resulting from the LSP on racial integration in the public sending schools and private receiving schools? In order to judge progress toward integration, we must first define a benchmark representing the level of integration a given school could reasonably achieve given the makeup of the surrounding community. For this benchmark, we use the racial composition of the larger metropolitan or micropolitan area, the Core Based Statistical Area (CBSA) as defined by the U.S. Census Bureau, in which the school is located. The CBSA is the most appropriate benchmark for the broader community, as it approximates the geographical area from which a school could reasonably be expected to draw students in the absence of legal or political boundaries. In total, 26 CBSAs are represented in our sample. The percentage of the school-age population that is white in these areas ranges from 27 percent to 74 percent, with an average value of 56 percent. The largest CBSA is the New Orleans-Metairie-Kenner metro area, which has a population of nearly 220,000.
We identify as integration-improving student transfers those that move a school’s racial composition closer to that of the relevant CBSA benchmark. Integration-reducing transfers are those that move a school’s racial composition further from the benchmark. For example, if an African American student leaves a school that is more African American than its surrounding community, we would say that this transfer has improved integration at the prior school. On the other hand, if a white student uses an LSP voucher to attend a school that is more white than its surrounding community, the transfer would be reducing integration at the new school. We classify transfers of all African American, Hispanic, and white students in this manner. This measure is well established for studying the integration impacts of school-choice programs.
We use the state’s student-level database to examine the voucher program’s impact on integration in the 2012–13 school year, the program’s initial year of statewide operation. First, we assess whether the private schools attended by students using state-funded vouchers offer more or less racially segregated environments than those available to students who remain in public schools. For this comparison, segregation is defined as the difference in each school’s percentage minority from that of the CBSA. Comparing private schools that have LSP enrollees to public schools that were previously attended by LSP users, we find that LSP schools are no more segregated, on average, than the public schools previously attended by voucher students. In addition, private schools with LSP enrollees are half as likely as public schools to be identified as racially homogeneous, which we define as having 90 percent of students belonging to the same race or ethnicity. Just 17 percent of LSP schools are racially homogeneous, compared to 34 percent of public schools that previously enrolled LSP students, a statistically significant difference.
Figure 1 summarizes our primary analysis of the effects of LSP transfers on integration in both sending and receiving schools. We identify transfers as improving integration if they bring a school’s racial proportions closer in line with those of the greater CBSA, and identify transfers as reducing integration if they bring the racial proportions further from those of the greater CBSA. As Figure 1 shows, among LSP transfers for which data are available, the overwhelming majority (83 percent) have positive impacts on the racial integration of the student’s sending school. While LSP transfers result in slightly more negative outcomes than positive ones for receiving schools, this difference is not statistically significant. Thus, while our analysis indicates that the use of LSP vouchers substantially improves integration for traditional public schools, the effect on receiving schools is insignificant.
When we examine the impact of transfers on sending schools by race, the majority of transfers by African American and Hispanic students improve integration in their former public schools. Fewer than one-third of white transfers are integration-improving. Nevertheless, because African American and Hispanic students make up nearly 90 percent of the analysis sample, the net effect on integration is positive.
Figure 2 looks at the impact of LSP transfers on public schools in the 34 districts that are under federal desegregation orders. When we restrict our analysis to these districts, we find that transfers improve integration in both sending and receiving schools. For sending schools, 74 percent of the moves improve integration, while just 26 percent have a negative impact. In receiving schools, 56 percent of total transfers improve integration. Again, when we break out our results by race, we find that nearly 90 percent of African American transfers improve desegregation in their sending schools.
These findings are essentially unchanged when we restrict our analysis further to only the 24 districts in which the U.S. is listed as a party in the original desegregation cases. In this restricted sample, LSP transfers improve integration in both sending and receiving schools, but the latter difference is not large enough to meet standard levels of statistical significance. Thus, it would appear that the voucher program is also beneficial for school integration in the very districts that are the subject of the Department of Justice litigation.
Our analysis of the Louisiana Scholarship Program reveals that the vouchers used by the subset of recipients for whom information is available have supported public-school desegregation efforts. By leaving schools in which their racial group was overrepresented relative to the surrounding communities, voucher users have improved integration in Louisiana public schools. At the same time, student transfers have, in general, no net negative impact on racial integration in their new schools. For African American students, who constitute the majority of voucher recipients, approximately 90 percent of LSP transfers improve integration for sending schools in both the overall sample and the subset of transfers in districts under desegregation orders. Based on this evidence, we conclude that the LSP is unlikely to have harmed desegregation efforts in Louisiana. To the contrary, the statewide school voucher program appears to have brought greater integration to Louisiana’s public schools.
Anna J. Egalite and Jonathan N. Mills are doctoral academy fellows in the Department of Education Reform at the University of Arkansas.