Deepening Segregation in American Public Schools

By Gary Orfield, Harvard University; Mark D. Bachmeier, David R. James and Tamela Eide, Indiana University Harvard Project on School Desegregation

Vol. 19, No. 2, 1997 pp. 11-18


After the Supreme Court outlawed segregated education in the South, it took fifteen years and a series of actions by the courts, Congress, the executive branch, and civil rights groups before the seventeen states with legal segregation were changed from an area of total educational segregation to the nation's most integrated. It remained that way for a generation. Now, there are clear signs that progress is coming undone and that the nation is headed backwards toward greater segregation of black students, particularly in the states with a history of de jure segregation.

The trends reported here are the first since the Supreme Court approved a return to segregated neighborhood schools under some conditions. A number of major cities have recently received court approval for such changes and others are in court. The segregation changes are most striking in the Southern and Border states but segregation is spreading across the nation, particularly affecting our rapidly growing Latino communities in the West. The racial and ethnic segregation of African-American and Latino students has produced a deepening isolation from middle class students and from successful schools. Our report also highlights a little noticed but extremely important expansion of segregation to the suburbs, particularly in larger metropolitan areas. Expanding segregation is a mark of a polarizing society without effective policies for building multiracial institutions.

Latino students, who will soon be the largest minority group in American public schools, were granted the right to desegregated education by the Supreme Court in 1973, but new data show they now are significantly more segregated than black students, with clear evidence of increasing isolation across the nation. In contrast to the varied regional trends and changes in direction over time for African Americans, Latino students are becoming

Page 12

more isolated almost everywhere. Part of this trend is caused by the very rapid growth in the number of Latino students in several major states. Regardless of the reasons, Latino students now experience more isolation form whites and more concentration in high poverty schools than any other group of students. This was long true in the centers of Puerto Rican settlement in the Northeast but it is rapidly increasing now for students in areas where the Latino communities are overwhelmingly of Mexican background.

The segregation is not simply racial separation, it is segregation by class and family and community educational background as well. Segregated black and Latino schools are fundamentally different from segregated white schools in terms of the background of the children and many things that relate to educational quality. Only a twentieth of the nation's segregated white schools face conditions of concentrated poverty among their children, but more than 80%, of segregated black and Latino school do. Desegregation is not only sitting next to someone of the other race. A child moving from a segregated African-American or Latino school to a white school will very likely exchange conditions of concentrated poverty for a middle class school. Exactly the opposite is true when a child is sent back from an interracial school to a segregated neighborhood school as is happening under a number of recent court orders which end busing or desegregation choice plans.

This is of fundamental importance to educational opportunity. The United States is a nation with a shrinking proportion of white students and a rising share of black and Latino students, groups which experience far less success in American public education and are concentrated in schools with lower achievement levels and less demanding competition. Recent court decisions approving a return to segregated neighborhood schools in various parts of the country will intensify the isolation.

The Supreme Court's 1954 conclusion that intentionally segregated schools are "inherently unequal," and contemporary evidence indicating that this remains true today means that it is very important to continuously monitor the extent to which the nation is realizing the promise of equal educational opportunity in schools that are now racially segregated. Education was vital to the success of black tenth of the U.S. population when de jure segregation was declared unconstitutional in 1954. It is far more important today, in an era in which millions of the good, low-education jobs have vanished. We are now talking about a society which has one-third non-white public schools and where whites will make up only half of

Page 13

the school age population in a third of a century if well-established trends continue. The stakes are much higher. We are moving backward toward greater separation rather than pressing gradually forward as we were between the 1950s and the mid-1980s for black students.

Our report presents the latest available evidence on segregation trends from federal enrollment statistics. It shows a delayed impact of the Reagan Administration campaign to reverse desegregation orders, which made no progress while Reagan was President but now has had a substantial impact through appointments which trans-formed the federal courts. The 1991-94 period following the Supreme Court's first decision authorizing resegregation witnessed the continuation of the largest backward movement toward segregation for blacks in the forty-three years since Brown v. Board of Education.

During the 1980s, the courts rejected efforts to terminate school desegregation and the level of desegregation increased, although the Reagan and Bush Administrations advocated reversals. Congress rejected proposals for major steps to reverse desegregation and there has been no trend toward increasing hostility to desegregation in public opinion. In fact, opinion is becoming more favorable. The policy changes have come from the courts. The Supreme Court, in decisions from 1991 to 1995, has given lower courts discretion to approve resegregation on a large scale and it is beginning to occur.

The statistics we report show only the first phase of what is likely to be an accelerating trend. These statistics for the 1994-95 school year do not reflect post-1994 decisions to end desegregation plans in a number of areas including metropolitan Wilmington, Broward County Florida, Denver, Buffalo, Mobile, Cleveland, and others. Important cases in a number of other cities are pending in court now.

Background Of Desegregation

Forty-three years ago, in 1954, the Supreme Court began the process of desegregating American public education in its landmark decision, Brown v. Board of Education. Thirty-three years ago, Congress took its most powerful action for school desegregation with the passage of the 1964 Civil Rights Act. Twenty-six years ago, in 1971, the great national battle over urban desegregation began with the Supreme Court's decision in the Charlotte, North Carolina busing case, Swann v. Charlotte-Mecklenburg Board of Education. With Swann, there was a comprehensive set of policies in place for massive desegregation in the South.

No similar body of law ever developed in the North and West. The Supreme Court first extended some desegregation requirements to the cities of the North and recognized the rights of Latino as well as black students from illegal segregation in 1973. In the early 1970s, Congress enacted legislation to help pay for the training and educational changes (but not the busing) needed to make desegregation more effective. The last major initiatives intended to foster desegregation took place more than two decades ago.

Since 1974, almost all of the policy changes have been negative and there has been a major increase in the nation's non-white population, particularly among school age children.

The key decision limiting metropolitan desegregation, Milliken v. Bradley, concerned metropolitan Detroit and was a drastic limitation on the possibility of substantial and lasting city-suburban school desegregation in what was rapidly becoming a society dominated by suburbia, a society in which only a small fraction of white middle class children were growing up in central cities. That decision ended significant movement toward less segregated schools and made desegregation virtually impossible in many metropolitan areas where the non-white population was concentrated in central cities.

The Supreme Court ruled that the courts could try to make segregated schools more equal in its second Detroit decision in 1977, Milliken v. Bradley II. The Court authorized an order that the State of Michigan pay for some needed programs in Detroit which were aimed at repairing the harms inflicted by segregation in schools that would remain segregated because of the 1974 decision blocking city-suburban desegregation. Unfortunately, there was little serious follow-up on the educational remedies by the courts and the Supreme Court would radically limit their reach in the 1995 Missouri v. Jenkins decision.

By far the most important changes in policy in the 1990s came from the Supreme Court. The appointment of justice Clarence Thomas in 1991 consolidated a majority favoring cutting back civil rights remedies requiring court-ordered changes in racial patterns. In the 1991 Board of Education of Oklahoma City v. Dowell decision, the Supreme Court ruled that a school district that had complied with its court order for several years could be allowed to return to segregated neighborhood schools. In the 1992 Freeman v. Pads decision, the Court made it easier to end student desegregation even when the other elements of a full desegregation order had never been accomplished. Finally, in its 1995 Jenkins decision, the Court's majority ruled that the court-ordered programs designed to make segregated schools more equal educationally and to increase the attractiveness of the schools to accomplish desegregation through voluntary choices were temporary and did not have to work before they

Page 14

could be discontinued.

In other words, desegregation was redefined from the goal of ending schools defined by race to a temporary and limited process that created no lasting rights and need not overcome the inequalities growing out of a segregated history. These decisions stimulated efforts in a number of cities to end the court orders, sometimes even over the objection of the school districts involved.

As the courts were cutting back on desegregation requirements the proportion of minority students in public schools was growing rapidly and becoming far more diverse. American public schools enrolled more than 43 million students in the fall of 1994 of whom 66% were white, 17% African American, 13% Latino, 4% Asian, and 1% Indian and Alaskan. By 1994, the proportion of Latinos in the U.S. was higher than that of blacks at the time desegregation began in 1954 and the proportion of whites far lower. The two regions with the largest enrollments, the South and the West, were 58% and 57% white, foreshadowing a near future in which large regions of the U.S. will have white minorities.

Rising African-American Segregation in the Heartland of the Old Segregation

The South and the Border State region are leading the nation in the turn back toward segregation for black students, because they have been the most desegregated region and have the most progress to lose. Ever since the civil rights revolution in the 1960s, the seventeen states of these two regions (the eleven states of the Old Confederacy and the adjoining six states from Oklahoma to Delaware which also maintained state-mandated segregation) have been the center of the least segregated region for black students. The transformation of this huge region, with more than one-third of the states, from an area of complete educational apartheid to the least segregated area in the U.S. was a historic accomplishment that is being lost.

Two of the three measures used in this study, show that the South has fallen behind another region of the country. The Border State region is now reporting an extremely high level of intense segregation, exceeded only by the Northeast. These regions are clearly slipping back toward their far more segregated pasts.

In terms of the proportion of black students in desegregated majority white schools, the South increased dramatically from virtually total segregation in 1960 to 14% of blacks in majority-white schools in 1967, 36% in 1972 and a high of 44% in 1983. Since then the number dropped to 39.2% in 1991 and 36.6% in 1994, losing all the slow progress of the last two decades and heading back toward the levels of segregation before the cities were desegregated. On the other measures of segregation the pattern for the region was similar. Its level of intense segregation increased slightly and the exposure of its black students to white students fell.

The Border State region, encompassing the six states from Oklahoma to Delaware which were not part of the old Confederacy but had a system of mandated segregation at the time of the Brown decision, experienced a more rapid rise in segregation from 1991-1995. The Border State region went from having 41% of its black students in majority white schools to 36% in just three years, a very rapid rate of change. The percent in intensely segregated schools climbed from 33% to 37% and exposure of black students to whites also declined significantly.

The most segregated regions for the past generation, the Northeast and the Midwest, continued to lead the list this year, except the Border States surpassed the Midwest in terms of intense segregation. Segregation in the most segregated region, the Northeast, remained about the same. The Northeast now has about half of its African-American students in schools that are 90-100% nonwhite, far surpassing other regions in the level of intense segregation.

Trends for Latino Students

Latino segregation has become substantially more severe than African-American segregation by each of the measures used in this study. In the Northeast, the West, and the South, more than three-fourths of all Latino students are in predominantly non-white schools, a level of isolation found for African-American students only in the Northeast. We have been reporting these trends continuously for two decades. They are clearly related to inferior education for Latino students. Though survey data are limited, the surveys that have been done tend to show considerable interest in desegregated education among the Latino families and substantial support for busing if there is no other way to achieve integration.

The data shows a continuing gradual national increase in segregation for Latino students. The most significant change comes in the proportion of students in intensely segregated schools, which rose to 34.8% in 1994. In 1968, only 23% of Latino students were in these isolated and highly impoverished schools compared to 64% of black students. Now the percentage of Latino students in such schools is up by almost half and is slightly higher than the level of intense segregation for black students.

Race and Poverty

The relationship between segregation by race and segregation by poverty in public schools across the nation is exceptionally strong. The correlation between the percent of black and Latino enrollments and the percent of students receiving free lunches is an extremely high 72. This means that when we talk about racially segregated schools, they are very likely to be segregated by poverty as well.

There is strong and consistent evidence from national and state data from across the U.S. as well as from other nations that high poverty schools usually have much lower levels of educational performance on nearly all outcomes. This is not all caused by the school: family background is a more powerful influence. Schools with concentrations of low income isolated children have less prepared children. Even better prepared children can be harmed academically if they are placed in a school with few other prepared students and, in some cases, in a setting where academic achievement is not supported.

School level educational achievement scores in many states and in the nation show a very strong relation between poverty concentrations and low achievement. This is because high poverty schools are unequal in many ways that effect educational outcomes. The students' parents are far less educated--a very powerful influence--and the child is much more likely to be living in a single parent home which is struggling with multiple problems. Children are much more likely to have serious developmental and untreated health problems. Children move much more then, often moving involuntarily in the midst of a school year, meaning that schools often do not have the students for sufficient time to make an impact. High poverty schools have to devote tar more time and resources to family and health crises, security, children who came to school not speaking standard English, seriously disturbed children, children with no educational materials in their homes, and many children with very weak educational preparation. These schools tend to draw less qualified teachers and to hold them for shorter periods of time. They tend to have to invest much more heavily in remediation and much less adequately in advanced and gifted classes and demanding materials. The level of competition and peer group support for educational achievement are much lower in high poverty schools. Such schools are viewed much more negatively in the community and by the schools and colleges at the next level of education as well as by potential employers. In states that implemented high stakes testing that denies graduation or flunks students, the high poverty schools tend to have the highest rates of sanctions by far.

None of this means that inc relationship between poverty and educational achievement is inexorable and that there are not exceptions. Many districts have one Or a handful of high poverty schools that performs well above the normal pattern. Students of the same family background may perform at ninny different levels of achievement, and there are some talented students and teachers in virtually every school the overall relationship, however, are very powerful. Students attending high poverty schools face a much lower level of competition regardless of their own interests and abilities.

This problem is intimately related to racial segregation. The data show that 60.7% of the schools in the U.S. have less than one-fifth black and Latino students while 9.2% have 80-100% black and Latino students. At the extreme, only 5.4% of the schools with 0-10% Black and Latino students have more than half low income students; 70% of them have less than one-fourth poor students. Among schools that are 90-100% African American and/or Latino on the other hand, almost nine-tenths (87.8%) are predominantly poor and only 3% have less than one-fourth poor children. A student in a segregated minority school is 16.3 times more likely to be in a concentrated poverty school than a student in a segregated white schools.

Segregation in the Suburbs

Blacks living in rural areas and in small and medium sized towns or the suburbs of small metropolitan areas are far more likely to experience substantial school de-segregation than those living in the nation's large cities. Students living in towns and rural areas and in suburbs of small metropolitan complexes attend school with an aver-age of about half white students. In contrast, those in the big central cities attend. schools those that have an aver-age of 83% nonwhite students. Suburbs of big and small central cities occupy an intermediate position, with black students in schools with about 40% whites and 60% non-white students. Considering the small proportion of minority students in many suburban rings this level of segregation is a poor omen for the future of suburbs which will become more diverse.

The nation's nonwhite population is extremely concentrated in metropolitan areas. Outside the South, this concentration tends to be in the largest metropolitan areas with the largest ghettos and barrios. Since the minority communities are constantly expanding along their boundaries, and virtually all-white developments are continuously being constructed on the outer periphery of suburbia, the central cities have a continual increase in their proportion of black and Latino students.

The suburbs are now the dominant element of our society and our politics. As the nation's population changes dramatically in the coming decades, they are destined to become much more diverse. What kind of access black and Latino children will have to mainstream suburban society will be affected by the racial characteristics of suburban schools. It raises serious concerns to realize that by 1994, blacks were in schools that averaged 59% nonwhite and Latino students were in schools what were 64% nonwhite in the suburbs of the largest cities although the whites in those suburban rings were in schools with little hope of stabilizing lists enrollment once a major racial change begins without drawing on students from a broader geographic area. This means that in areas with many fragmented school districts, not only the central city but also substantial portions of suburban rings may face high levels of segregation.

Since nonwhite suburbanization began in earnest in the 1970s, the cities have also been losing many of their minority middle class families, leaving the central cities with a higher and higher concentration of poverty. These districts contain 18% of the black students, 23% of the nation's Latino students, 13% of Asians, but only 2% of the whites. About a fifth of black and Latino students depend an average of only 14% combined black and Latino enrollments. Latino students, but not blacks, were almost as segregated in the suburbs of smaller metropolitan areas. If these patterns intensify as the suburban African-American and Latino population grows, we may be facing problems that are as serious as those that led to desegregation conflicts in many central cities.

Table 1: Most Segregated States for Black Students, 1994-1995 on Three Measures of Segregation

%of Blacks in Majority White Schools %of Blacks in 90-100% Minority Schools Average % of White in Schools of Typical Black Student
New York 15.1 Illinois 61.9 Illinois 20.0
California 17.5 Michigan 59.6 New York 20.1
Michigan 18.8 New York 57.1 Michigan 20.8
Illinois 20.2 New Jersey 53.7 New Jersey 25.7
Mississippi 23.3 Pennsylvania 47.0 California 26.0
Maryland 25.9 Maryland 46.0 Maryland 27.3
New Jersey 26.6 Alabama 38.2 Mississippi 28.5
Louisiana 28.8 Tennessee 38.0 Louisiana 30.4
Wisconsin 29.2 Louisiana 37.6 Pennsylvania 30.4
Pennsylvania 30.4 Mississippi 36.9 Alabama 32.9

It would be profoundly ironic if the Supreme Court decision that meant to protect suburban boundary lines, Milliken v. Bradley, ended up making it impossible for suburban communities in the path of racial change to avoid rapid resegregation. Individual suburban school districts are often so small that they can go though racial change much more rapidly and irreversibly than a huge central city. A suburb will often have only the enrollment of a single high school attendance area in a city and has little hope of stablizing lists enrollment once a major racial change begins without drawing on students from a broader geographic area. This means that in areas with many fragmented school districts, not only the central city but also substantial proportions of suburban rings may face high levels of segregation.

Since nonwhite suburbanization began in earnest in the 1970's, the cities have also been losing many of their minority middle class families, leaving the central cities with a higher and higher concentration of poverty. These districts contain 18% of the black students, 23% of the nation's Latino student, 13% of Asians, but only 2% of the whites. About a fifth of black and Latino students depend on districts that do not matter to 98% of white families. Most of these systems have faced recurrent fiscal and political crises for years and have low levels of educational achievement. Desegregation has become virtually impossible in some of these systems since the 1974 Supreme Court decision in the Detroit case. The trends of metropolitan racial change that have been operating since World War II suggest that segregation will become worse in the future.

Segregation of Whites

In a nation where whites are destined to become one of several minorities in the schools if the existing trends continue, it is important not only to consider isolation of nonwhite students from whites but also the isolation of whites from the growing parts of the population.

Page 17

Except in the historic de jure states for blacks and the states taken from Mexico in the war in the 1840s, most white students have not yet experienced substantial desegregation. Although they are growing up in a society where the Census Bureau predicts that more than half of school age children will be non-white in a third of a century, many are being educated in overwhelming white schools with little contract with black or Latino students. As the American workforce changes skills in race relations will become increasingly valuable in many jobs.

Southern and Border states tend to have substantially larger proportions of black students than other states and higher levels of desegregation. Not only did they have to overcome a very rigid segregation but they also had to bring much larger shares of black students into interracial schools. As a result of desegregation orders their white students tend to be in schools with much larger proportions of black students than whites in other parts of the country. In no state in the Northeast, the Midwest, nor the West do whites, on average, go to schools with 10% of more black classmates. Thirteen Southern and border states have higher shares of black students than any state in the North or West. White students attend school with the largest proportions of black students in the states indicated in Table 2.

The most segregated states of the North had little interracial experience for white students. In Illinois, New York, and California the typical white student attended a school with less than 7% blacks; in New Jersey, less than 8%; and in Michigan less than 5%. Some states with developing patterns of serious segregation, such as Wisconsin and Minnesota, have proportions that are even lower.

There were twenty-three states where the number of black students per classroom in the schools attended by white students was less than 1 percent. Northern whites who may be inclined to assume their superiority to Southerners in terms of racial attitudes and policies should reflect on the fact that Southern white students have experienced far more actual desegregation for a quarter century and some northern states with very small black enrollments have been more resistant and made less progress than Southern states with far larger black communities.

The exposure of whites to schools which average more than 10% Latino enrollment is limited to six South-western states--New Mexico (36%), California (23%), Texas (19%), Arizona (19%), Colorado (13%), and Nevada (13%). In eight other states, the typical white is in a school with more than one-twentieth Latinos. The number is lower in the other thirty-six states and that is a major reason that national attention on this extremely important population is limited. In the long run, secondary immigration may distribute Latinos much more broadly across the nation.

Federal and State Policy

The Clinton administration has no stated policy on the movement back toward segregation and has given no priority to supporting successful desegregation. Although it is presiding over the period of the most rapid resegregation of the South since the Brown decision, it has not proposed any initiative, though the hostility of the previous twelve years has ended and positions have been changed on some important cases. There has been no proposal to restore the federal desegregation aid program that reached its peak under President Carter and whose funding was eliminated under President Reagan. Although the administration is asking for large increases in compensatory education, to a total of $7 billion for children in high-poverty, low performance schools, it has no initiative to move children out of such failing schools or even to slow the termination of desegregation plans in communities where equal education for minority students has never been achieved.

The only two desegregation-related items left in the education budget are for the small desegregation assistance centers set up under the 1964 Civil Rights Act and the Magnet school program, which is often used to help pay for some of the voluntary elements in desegregation plans. The White House has requested zero increase for each program.

School desegregation has not been chosen as a priority issue by the Education Department's Office for Civil Rights or the Justice Department's Civil Rights Division, and no major new research on the consequences of segregation or the best methods for improving the successful operation of multiracial schools and classrooms have been commissioned. The leader of both the Justice Department's Civil Rights Division and the Education Department's Office for Civil Rights recently stated that neither department has issued any statement of its policy on school desegregation during the Clinton Administration and that each is responding on a case-by-case basis.

Very few states have active state government efforts to enforce school desegregation in the middle 1990s. Some that once had rules or state legal requirements have suspended them or have terminated the offices that administered them. In California, for example, where segregation was increasing rapidly for both blacks and Latinos, and for some groups of Asian students, the State Department of Education's Intergroup Relations Office was abolished, though the state provides funds for court ordered remedies. In Illinois, the state supreme court took away the state board of education's right to enforce

Page 18

desegregation requirements. In New Jersey, the state desegregation office has been abolished under Governor Christie Whitman.

Table 2: States with Lowest Levels of Segregation of Whites

State Average % black in School of Typical white
Mississippi 30.3%
South Carolina 29.9%
Delaware 27.2%
Louisiana 26.8%
North Carolina 22.9%
Georgia 21.6%
Alabama 18.9%
Virginia 17.7%
Florida 16.9%
Maryland 15.9%

Many states have adopted policies to publicize achievement results by district and school and they repeatedly publish lists that show urban minority schools with very high levels of concentrated poverty at the bottom in academic achievement without ever discussing the very frequent relationship between segregated education and low achievement. If standards are to be raised with high stakes for students, states must be concerned about the structural fairness of their system to minority students.

The Bridge Backward

In American race relations, the bridge from the twentieth century may be leading back into the nineteenth century. We may be deciding to bet the future of the country once more on separate but equal. There is no evidence that separate but equal today works any more than it did a century ago.

The debate that has been stimulated by recent Supreme Court decisions is a debate about how and when to end desegregation plans. The most basic need now is for a serious national examination of the cost of resegregation and the alternative solutions to problems with existing desegregation plans. Very few Americans prefer segregation and most believe that desegregation has had considerable value, but most whites are still opposed to plans which involve mandatory transportation of students. During the last fifteen years plans have been evolving to include more educational reforms and choice mechanisms to try to achieve desegregation and educational gains simultaneously. A stronger fair housing law, a number of settlements of housing segregation cases, and federal initiatives to change the operation of subsidized housing--as well as the very rapid creation of brand new communities in the sunbelt--all offer opportunities to try to change the pattern of segregated housing that underlies school segregation.

Policies that would help move the country back toward a less polarized society include:

1) resumption of serious enforcement of desegregation by the Justice Department and serious investigation of the degree to which districts have complied with all Supreme Court requirement by the Department of Education. Such requirements could be appropriately specified in a federal regulation.

2) creation of a new federal education program to train students, teachers, and administrators in human relations, conflict resolution, and multi-ethnic education techniques and to help districts devise appropriate plans and curricula for successful multi-racial schools.

3) serious federal research on multiracial schools and the comparative success of segregated and desegregated schools.

4) a major campaign to increase nonwhite teachers and administrators through a combination of employment discrimination enforcement and resources for recruitment and education of potential teachers.

5) incorporation of successful desegregation into the national educational goals.

6) federal and state efforts to expand the use of integrated two-way bilingual programs from the demonstration stage to become a major technique for improving both second language acquisition for both English speakers and other language speakers and successful ethnic relationships.

7) additional Title W resources to expand state education department staffs working on desegregation and racial equity in the schools.

8) federal, state, and local plans to coordinate housing policy with school desegregation policy.

9) examination of choice and charter school plans to assure that they are not increasing segregation and to reinforce their potential contribution to desegregation.

10) examination of high stakes state testing programs to assure that they are not punishing the minority students who must attend inferior segregated schools under existing state and local policies.