City Lines, County Lines, Color Lines: The Relationship between School and Housing Segregation in Four Southern Metro Areas

by Genevieve Siegel-Hawley - 2013

Background/Context: At the close of the first decade of the 21st century, the intersection of race, geography and opportunity is increasingly referred to as spatial racism. School quality and resources, municipal services, employment opportunities, accessibility of transportation, exposure to pollution, and tax rates all vary dramatically across a network of invisible boundary lines that carve up U.S. metro areas into racially and socioeconomically distinctive spaces.

Purpose/Objective/Research Question/Focus of Study:This analysis explores how district boundary lines and school desegregation policy have impacted metropolitan school and housing integration levels over the past two decades.

Setting: The study focuses specifically on the South, where the most comprehensive desegregation strategies were pursued. Based on varying experiences with city-suburban school district mergers, four metros—Richmond, Virginia; Charlotte-Mecklenburg County, North Carolina; Louisville-Jefferson County, Kentucky; and Chattanooga-Hamilton County, Tennessee—were identified for study between the years 1990 and 2010. This time period encompassed a rising trend of unitary status for school districts, the serious resegregation of students, and growing demographic complexity across the country.

Research Design: This is a quantitative analysis of school and housing segregation trends that relies upon Geographic Information Systems (GIS) maps to depict spatial patterns of isolation. School enrollment and segregation trends are derived from the National Center for Educational Statistics’ Common Core of Data. Housing patterns describing the distribution and segregation of persons by race and poverty status are explored using block group-level information from the U.S. Census Bureau’s decennial census. Segregation is measured using the Index of Dissimilarity (D).

Findings/Results: Efforts to overcome the divisive nature of district boundary lines, in conjunction with comprehensive school desegregation policy, were related to unambiguous progress in combating both school and housing segregation. This central finding suggests that, in some ways, school policy can become housing policy.

Conclusions/Recommendations: How we provide equal educational opportunity to an increasingly diverse population of students is one of the principal concerns facing the country today. As a nation, however, we are currently allowing a labyrinthine system of school-district boundary lines to dictate the geographic distribution of learning opportunities. This study provides concrete examples of metros implementing a more regional approach to educational equity. It recommends that other metro area consider taking similar steps to subvert the segregating power of school-district boundaries.

The problem of the Twentieth Century is the problem of the color-line.

W.E.B. Dubois, The Souls of Black Folk, 1903

A child's educational opportunity should be determined by her intellect and work ethic, not by her zip code.

Virginia Governor Bob McDonnell, GOP Response to President

    Barack Obamas State of the Union, January 2010

Two pronouncements uttered more than one hundred years apart, from opposite sides of the racial divide, further separated by an ideological chasm, concur: lines matter. The contours of those lines, parsed into geographical shorthand by Gov. McDonnell, may look somewhat different today than when Dubois penned the opening paragraph of The Souls of Black Folk, but the fundamental issues remain the same. At present, a network of invisible boundary lines fracture school districts and municipalities, giving structure to low-opportunity areas of racial and socioeconomic isolation (Briggs, 2005; G. Orfield & McArdle, 2006; Weiher, 1992). And it is largely because of school district boundaries that a childs educational prospectsand, consequently, his or her future opportunitiesare indeed intertwined with his or her zip code.

For children, opportunity depends in large part on the neighborhood and schools they will grow up and learn within. Those connected to high-opportunity geographic spacesreplete with good housing, strong schools, favorable environmental conditions, quality health care, and dependable public servescome into a very different set of life chances than children locked into segregated, low-opportunity areas (Jargowsky & El Komi, 2011; Lareau, 2003). These early experiences carry deep consequences for upward mobility, particularly in the form of educational and job prospects.

More than four decades ago, the nations highest court stood poised to alleviate the segregating effects, and related implications for opportunity, of school-district boundary lines. Swann v. Charlotte-Mecklenburg, a 1971 case out of Charlotte, North Carolina, signaled that the Supreme Court recognized the critical link between school and housing segregation. Attempting to disentangle that relationship and holding that desegregation must be achieved to the greatest extent possible, the justices ordered the two-way busing of Black and White students across the already merged city-suburban school system in Charlotte-Mecklenburg (G. Orfield & Eaton, 1996; Swann v. Charlotte-Mecklenburg, 1971). A looming question remained, however: How would comprehensive school desegregation be carried out in metropolitan areas where the city-suburban lines remained intact?

In 1974, a Supreme Court packed with Nixon appointees agreed to take up that question. In Milliken v. Bradley, the newly rightward-leaning court discounted evidence of the governments past involvement in contributing to racial isolation and ruled against a metropolitan desegregation remedy for Detroit. The decision meant that, barring proof of intentional suburban discrimination, future desegregation efforts occurring across district boundary lines would have to be voluntary. Milliken thus hardened the linesand the inequitiesbetween cities and suburbs.

Central to this analysis, though, is the way in which the splintering effects of the 1974 Milliken decision have been ameliorated in the South. Along with a handful of court-ordered metropolitan mergers, several states in the region, like North Carolina and Tennessee, operate under laws that facilitate city-suburban school consolidation. These regional characteristics created an ideal backdrop for a study examining the long-term school and housing segregation trends associated with metropolitan school districts.

Based on widely differing histories of city-suburban school district mergers, four metrosRichmond, Virginia; Charlotte-Mecklenburg County, North Carolina; Louisville-Jefferson County, Kentucky; and Chattanooga-Hamilton County, Tennesseewere identified for study between the years 1990 and 2010. These metros are part of a region that is one of the most racially diverse areas in the nation, where the former Black-White paradigm has shifted to a situation where one in five students is Latino (G. Orfield, 2009). The South is also the first area of the country to report a majority share of low-income students (Southern Education Foundation, 2010). Moreover, after widespread desegregation during the 70s and 80s, since the early 1990s, many southern school systems have experienced rapid resegregation (Boger & Orfield, 2005; G. Orfield & Eaton, 1996). The rise of this new Southmore diverse and yet containing schools that are rapidly becoming more segregatedraises critical questions about the future provision of equitable educational opportunities in cities and schools across the region.  

Using cutting-edge spatial and graphic imagery and a multi-disciplinary framework, this analysis adds to the body of evidence related to school desegregation experiences across the South. In the region where the most comprehensive strategies were pursued, This article explores how contemporary (1990-2010) school and housing segregation patterns have been impacted by desegregation policy and the manipulation of district boundary lines in four southern metropolitan areas. Related areas of inquiry include the following: First, how are patterns of school and housing segregation affected by different types of desegregation policy (e.g. controlled choice versus magnet schools)? Second, in what way do school and housing segregation patterns shift after desegregation plans are discarded?

Findings indicated that efforts to overcome the divisive nature of district boundaries, in conjunction with school desegregation policy, were related to unambiguous progress in combating school and housing segregation. Indeed, results revealed that, in some ways, school policy can be housing policy: metros with city-suburban school desegregation plans reported much faster declines in housing segregation than metros lacking such strategies. In a nation experiencing swift demographic transition and rising inequality, evidence from these four southern locales emphasizes the ongoing and urgent need for policies explicitly designed to equalize educational and life opportunities.

The first section of this article presents background information related to the impact of metropolitan fragmentation, the relationship between school and housing segregation, and the power of metro-wide school desegregation policy. The second section delves into key characteristics of each of the sites, followed by a description of the data and methodology. Next, the article provides an analysis of school and housing segregation in and across the four metro areas. It closes with a discussion of key findings, directions for future policy, and a conclusion.

BACKGROUND

METROPOLITAN FRAGMENTATION AND ITS CONSEQUENCES

Recent estimates show that the United States contains nearly 90,000 units of government (U.S. Census Bureau, 2010). Excluding the 51 entities that make up the federal and state governments, local agencies accounted for fully 89,476 units, making the U.S. one of the most fragmented political systems in the world (Weiher, 1991). The vast majority of these local governments, and their associated boundaries, are concentrated in American metropolitan areas (M. Orfield & Luce, 2010).

Metro area divisions matter greatly on any number of levels. School quality and resources, municipal services, employment opportunities, accessibility of transportation, exposure to pollution, and tax rates all vary dramatically across the multiple boundary lines that carve up U.S. metros into racially and socioeconomically distinctive spaces (M. Orfield & Luce, 2010; Powell, 2009; Rusk, 1999).

Research specifically delineates the role boundaries play in exacerbating school segregation and limiting educational opportunity (Clotfelter, 2004, Eaton, 2001; Grant, 2009; Reardon & Yun, 2002; Ryan, 2010). One recent study, for example, used census and school enrollment data to analyze the supposedly race-neutral creation of numerous new school districts in Jefferson County, Alabama. The author found that the existence of these newly fragmented school systems greatly aggravated school segregation across district boundaries (Frankenberg, 2009). Another recent research project used school observations and in-depth interviews with school district officials in hyper-fragmented Long Island to uncover tremendous resource disparities between racially and socioeconomically isolated school systems and more privileged ones (Wells et al., 2009).

Disparities across municipal and school-district boundary lines exist largely because, without an overarching political body managing regional concerns, small local units compete with one another for advantageseconomic or otherwise. For instance, developing suburbs are incentivized to build steeply-priced single family homes on large lots to encourage low population density (and thus fewer costly municipal service needs) and higher property taxes (Freund, 2007; M. Orfield & Luce, 2010). Some scholars have suggested that the competition between locales is natural and beneficial to metropolitan areas, effectively sorting the population according to their preference for government services and tax rates (Beck, 1983; Tiebout, 1956). This model assumes, of course, that everyone in a metropolitan area has full information about the availability of other options, the financial means to relocate and faces no discrimination in the housing market. None of these assumptions actually hold true in the American metropolis (National Commission on Fair Housing and Equal Opportunity, 2008; Shapiro, 2004; Turner & Ross, 2003).

THE COMPLEX RELATIONSHIP BETWEEN BOUNDARY LINES AND SCHOOL AND HOUSING SEGREGATION

The common practice of drawing school-attendance boundaries to encircle nearby neighborhoods buttresses a reciprocal and cyclical relationship between school and housing segregation (Mickelson, 2011; Sohoni & Saporito, 2009). Because attendance zones concretely define the neighborhoods with which schools are associated, disentangling residential decisions from schooling preferences is extremely difficult. These invisible boundary lines help structure patterns of segregation by ensuring that racial or economic isolation present in neighborhoods will be reflected in the school enrollment. On a larger scale, district lines send demographic signals about entire systems of education. Families with children moving to or across the typical metro thus face a series of racialized decisions between attendance zones and school districts. And while those choices are often billed as free and open, for many families they are constrained by race and/or income level.

Choosing Neighborhoods and Schools

Families with resources talk about the ability to buy into a better school zone or district, based on the assumption that a wealthy neighborhood means good schools (Holme, 2002; Johnson, 2006; Shapiro, 2004). The reasoning behind this assumption is largely accurate. Property taxes that continue to make up a significant portion of local education funds (Kozol, 1992; Ryan, 2010) are related to home values, and housing prices in turn are linked to school districts and school zones (Chiodo, Hernández-Murillo, & Owyang; Kane, Reigg, & Steger, 2010). These circumstances bolster the relationship between educational and housing choices, making it difficult for lower-income families to gain entrance into highly resourced schools.

In addition to the limitation that socioeconomic status places on a familys ability to choose a neighborhood and school, racial disparities related to the accumulation of wealth in America also help define those choices (Di & Liu, 2005; Masnick, 2001; Oliver & Shapiro, 1995). The wealth gap translates into differential rates of homeownership for racial groups and, relatedly, differential racial access to wealthy neighborhoods and schools (Shapiro, 2004). Documenting one of the sources of disparate access, a series of interviews with more than 200 families in several large American metros uncovered an extensive web of family assistance for young, lower-middle-class White families (Johnson, 2006). Grandparents, for example, would often help place a down paymentsometimes covering it outrighton a home in a school district that would otherwise be out of reach for their children and grandchildren (Johnson, 2006). For families of color lacking those networks and resources, opportunities for homeownership in an expensive neighborhood with well-funded schools are more limited.

School Demographics and Achievement Send Signals to Prospective Homebuyers

Qualitative studies have attempted to dissect the underlying motivations behind family moving decisions and have found that interviewees speak knowledgeably about their housing options in relationship to school districts, even down to specific schools within the district (Holme, 2002; Shapiro, 2004). In some cases, White home-seekers are guided by perceptions of school quality unrelated to tangible educational characteristics, relying instead upon racially coded pieces of information shared among social networks.  One UCLA researcher conducted fieldwork on White families moving for the schools in a Southern California school district. She found that informal conversations passed through friends about the degree to which educational settings were serving whiter and wealthier students largely formed the basis for decisions about whether or not a zone or district should be sought out (Holme, 2002).

Indeed, race and racism continue to play a central role in determining access to predominately White and wealthy neighborhoods and schools, even as our post-racial society has become less willing to acknowledge persistent patterns of discrimination. One reason for these denials may be related to the subversion of blatant racial hostilities, which, in many cases, have been replaced with less obtrusivebut no less damagingbarriers to equal access to opportunity-rich neighborhoods. Research based on an intensive historical case study of Detroit indicated that White homeowners became much less likely to express overt racial prejudice in the decades following World War II, relying instead on an ideology linking together home-buying, property values, and racial exclusion (Freund, 2007). This new dialogue around "race, property and neighborhood integrity" has allowed Whites to shroud racist attitudes beneath the coded language of protecting home investments (Freund, 2007, p. 10).

Given these evolving politics around race and homeownership, it makes sense that the racial makeup of a school district appears to impact real estate prices. A Connecticut study based on data from 1994 to 2004 found that housing prices varied dramatically according to the percentage of Latino students in a school district. After controlling for a variety of neighborhood variables, the researchers found that higher home prices were reported in districts with fewer Latino students (Clapp, Nanda, & Ross, 2008). Using similar data from Connecticut, a second study found that buyers were willing to pay $7,468 more for a house in order to live near a less diverse school (Dougherty et al., 2009). Further south, North Carolina research examining the Charlotte housing market after the district returned to neighborhood school assignments found that housing prices varied significantly on either side of school-attendance boundary lines (Kane et al., 2010). Evidence from two different regions, then, suggests that home prices can be closely connected to the racial makeup of a school zone or district.

Narrow achievement indicators like test scores may herald a new way of discussing schools and neighborhoods in a coded manner. Certainly, a byproduct of the recent policy focus on testing and accountability has been the parental capacity to distill the complicated elements that come together to produce a school environment down to a single numerical score. The Connecticut-based research discussed above provides evidence in support of that trend, finding that test scores became increasingly predictive of housing prices after the passage of No Child Left Behind in 2001 (Clapp, Nanda & Ross, 2008).

A recent Brookings Institute study further illuminated the relationship between test scores and home prices by examining how the two are related to land-use policy. Using a variety of different data sources, the author found that in the 100 largest U.S. metropolitan areas, housing prices near a high-scoring public school are nearly 2.4 times greater (nearly $11,000 more per year), on average, than those near a low-scoring public school (Rothwell, 2012). These price differences contributed to a situation where the typical low-income student went to a school scoring at a much lower percentile on state tests than did the typical middle- or high-income student. Since exclusionary zoning practices effectively block low-income families entrance into the highest-performing schools, the study concludes that land-use policies have a tremendous impact on educational opportunity.

Real Estate Agents Use Schools to Steer Clients

Research has documented extensive real estate practices geared towards providing prospective buyers with racialized information about local schools. Because providing direct information about the racial composition of a neighborhood is typically illegal (USHD, 1979), housing agents often discuss school demographics or quality (which, as we saw, is often conflated with racial composition or test score results) with prospective buyers. One analysis found that real estate agents in racially fragmented metro areas were more likely to place ads identifying schools by name. The author of the study surmised that the name of the school served as a code for, this house is in a white neighborhood with a white school (Pearce, 1980, p. 14). A more recent probe of real estate sales found that 87% of testers were steered into specific neighborhoods, with agents using the racial composition of school districts as a proxy for neighborhood demographics (National Fair Housing Alliance , 2006).

Interviewers in hyper-fragmented Long Island, New York have also found that the relationship works in the other direction, with some school officials hosting meetings with realtors to tout the quality of schools in particular districts. These efforts provide another concrete example of close relationships between the parties who help intimately link school and housing decisions (Wells et al., 2009).

In sum, housing segregation remains closely connected to school segregation. Family choices based on racial and socioeconomic signals, test scores, coded language, and real estate practices help perpetuate a cycle of segregation where racially and socioeconomically identifiable schools yield racially and socioeconomically identifiable neighborhoods, and vice versa. Such a cycle is very difficult to disentangle without comprehensive policy designed to undermine it (Powell & Kay, 2001).

IMPACT OF SCHOOL DESEGREGATION POLICY

The yellow school bus continues to be one of the most definingand controversial symbols of school desegregation. Transportation is fundamental to desegregation policy because of the way boundary lines have shaped residential and school segregation patterns. Used to transcend both school zones and district boundaries, buses have traditionally helped break the link between housing and school segregation.

Relying heavily on transportation, a number of different school desegregation strategies have been introduced over the years. They run the gamut from early, mandatory student assignment policies to the rise of voluntary integration mechanisms like magnet schools and controlled choice (G. Orfield & Eaton, 1996). During the 1970s, metropolitan-wide school desegregation strategies implemented in a handful of metro areas under different conditions gained particular favor due to their ability to both stem long-standing patterns of White flight to the suburbs and to promote stable, integrated schools and neighborhoods (Frankenberg, 2005; G. Orfield, 1978; Parents Involved v. Seattle, 2006; Pearce, 1980; U.S. Commission on Civil Rights, 1977).

Indeed, earlier research shows that districts with comprehensive metro school desegregation plansincluding city-suburban transfers and magnet schoolshave the most stably integrated schools (Frankenberg, 2005; G. Orfield, 2001; M. Orfield, 2006; Parents Involved v. Seattle, 2006). Past studies also show that, between 1970 and 1990, regions with metro-wide school desegregation plans experienced decreases in residential segregation at more than twice the national average (Frankenberg, 2005; Parents Involved v. Seattle, 2006).

DYNAMICS OF METROPOLITAN SCHOOL DESEGREGATIONS POSITIVE INFLUENCE ON SCHOOL AND HOUSING PATTERNS

The success of metropolitan-wide desegregation efforts in reducing both school and housing segregation has a theoretical basis in the founding documents of this country. In 1787, James Madison, writing in heated support of the ratification of the Constitution (and thus weighing in on the side of a strong, unified federal government), warned against the dangers of factionalism, which he defined as a number of citizens&united and actuated by some common impulse of passion&adverse to the rights of other citizens, or to the permanent and aggregate interests of the community (1787, p. 1). The remedy to this type of dissent, he concluded, was to extend the sphere&take in a greater variety of parties and interests&make it less probable that a majority of the whole will have a common motive to invade the rights of other citizens (p. 5). Madisons Federalist Paper No. 10 thus outlined a basic but key governmental principle: The most effective way to combat the pursuit of narrow political interests is to broaden the limits of the community itself.

Metropolitan school desegregation is an important example of Madisons theory. Metro-wide desegregation plans are designed to extend across an expansive city-suburban community. Once the boundary lines separating city school districts from suburban ones are bridged, the constricted self-interests of White and/or middle class families seeking more homogenous, higher-scoring schools linked to higher home property values can be subverted. Schools across the metro area can then operate in service of a broader ideal that aims for a unified, integrated, high-quality educational system benefitting all members of the community (G. Orfield, 2001).

Another way of thinking about the mechanism behind the success of metropolitan school desegregation is to imagine the housing market in a fragmented metropolitan area where racially identifiable school systemseither predominately White or predominately minorityare located on either side of a school district line. In that metro area, families and real estate agents can easily communicate about school preferences between the districts without ever having to mention race. These conscious or unconscious maneuvers can occur even though the real estate decisions will, of course, have serious racial implications for both schools and neighborhoods (Pearce, 1980). In a different metro, one where comprehensive city-suburban school desegregation is carried out in a predictable fashion, the community comes to understand that movement to any part of the metro area governed by the student assignment policy will result in a similarly integrated schooland thus connected to the kinds of rich benefits that flow from those environments (see, e.g., Linn & Welner, 2007; G. Orfield, Frankenberg & Garces, 2008).

Integrated school settings are linked to both short-term and long-term benefits for students of all races. The more immediate associated outcomes include greater academic achievement, heightened critical thinking skills, an amplified ability to communicate and make friends across races, and a reduction in students willingness to accept stereotypes (Hawley, 2007; Mickelson & Bottia, 2010; Pettigrew & Trop, 2006; Ready & Silander, 2011; Schofield, 1995). These trends later translate into loftier educational and career expectations (Crain, 1970; Dawkins, 1983; Kurlaender & Yun, 2005) and high levels of civic and communal responsibility (Braddock, 2009).

Most important, perhaps, is the finding that graduates of racially diverse schools are more likely to attend integrated colleges and live in racially diverse neighborhoods (Braddock, 1980, 2009; Butler, 2010; Stearns, 2010; Wells & Crain, 1994). The reinforcing and intergenerational nature of this cycle (see Figure 1)whereby products of integrated K-12 settings are more likely to seek out integrated neighborhoods, making school integration more likely for their children (Mickelson, 2011)supplements Madisons well-articulated rationale for a robust federal government by helping to further illuminate how a long-standing metropolitan school desegregation plan might eventually produce advances in both school and housing integration.

Figure 1. Perpetuating effects of school desegregation

Note. Reprinted from The Reciprocal Relationship between integrated schooling and integrated housing: A synthesis of social science evidence, by R. Mickelson, 2011, in Finding Common Ground: Coordinating Housing and Education Policy to Promote Integration, Washington, DC: Poverty & Race Research Action Council.

It is likely safe to say that all of these different dynamics interact to produce the mechanism by which school desegregation policy might become housing desegregation policy. This analysis will put Madisons philosophy and prior research and theory to a contemporary test, seeking to understand whether metropolitan school desegregation remedies continue to promote school and housing integration across broadly conceived communities.

SITE SELECTION

A school systems capacity to override the splintering impact of district boundary lines was central to this analysis. Two of the four metros selected, Charlotte-Mecklenburg County and Chattanooga-Hamilton County, are located in states governed by statutes supporting a key principle: one county, one school system (unlike other states where multiple, fragmented school districts might exist within county lines). School districts in North Carolina and Tennessee thus have several methods at their disposal when seeking to consolidate separate systems.  

On the other hand, state laws and policies in Virginia and Kentucky do not easily lend themselvesunder current parameters, at leastto school system mergers. Instead, during the 1970s, districts in the Louisville and Richmond areas came under court order to consolidate as part of metropolitan-wide desegregation decrees. The experiences of the two metros diverged sharply, however, after higher courts overturned a district courts plan to desegregate Richmond-area students across a consolidated city-suburban school system. Instead, Richmond City Schools implemented its desegregation plan within the city limits, while the school systems in neighboring Henrico and Chesterfield counties remained exempt (Ryan, 2010). Louisville-Jefferson County, by contrast, carried out a 1975 merger and continues to implement a student assignment plan dedicated to promoting racially diverse schools across the metropolitan school district (G. Orfield & Frankenberg, 2011).  

These differing consolidation origins meant that school desegregation was a central focus and thrust of the court-ordered mergers in the Richmond and Louisville areas and merely an underlying consideration for some stakeholders in the Charlotte and Chattanooga cases (see Table 1).

Table 1. School Boundary-Lines and Desegregation Policy, Four Southern Metropolitan Areas

	Louisville-Jefferson County, Kentucky	Charlotte-Mecklenburg County, North Carolina	Chattanooga-Hamilton County, Tennessee	Richmond-Henrico-Chesterfield, Virginia
Boundary-line configuration	Merged city-suburban school district	Merged city-suburban school district	Merged city-suburban school district	Failed merger; city school district distinct from two surrounding suburban districts
Origins of merger	Court-ordered merger of city and suburban school systems (1974); formalized by the Kentucky Board of Education (1975)	Residents voted by a 3-1 majority to merge the city school system with Mecklenburg County schools (1959)	City school system ceased operations after narrowly approved referendum; folded into Hamilton County Schools (1997)	Supreme Court allowed lower court ruling rejecting school district consolidation to stand (1973)
Contemporary student assignment policy	Controlled choice policy governs student assignment plan; plan based on geographic distribution of students by race and poverty	Choice-based  policy prioritizing neighborhood schools and system of magnet programs	System of magnet programs, otherwise student assignment prioritizes neighborhood proximity	Open enrollment in Richmond City; assignment prioritizing neighborhood proximity in Henrico and Chesterfield

Sources: Bradley v. Richmond, 1973; Douglas, 1995; Eichenthal & Windeknecht, 2004; Hamilton County School District, 2011; Keller, 2006; Newburg v. Area Council, 1971; Mickelson et al., 2009; Phillips et al., 2009; Pratt, 1989; Ryan, 2010; Smith, 2004; Timeline: Desegregation in Jefferson County Public Schools, 2005.

In addition to the various boundary-line configurations, contemporary student assignment policywhich, in the case of the Louisville and Chattanooga areas, also represents current desegregation effortsdiverges widely across the four metros. Each falls along a continuum of policy designed to promote the creation of diverse schools across a metropolitan area. At one end of that policy spectrum, the Louisville area school system remains the only district under study that continues to pursue a widescale, voluntary integration strategy. The metros commitment to promoting diverse schooling experiences has continued despite recent political and legal challenges: A recent survey found that nearly 90% of families supported district guidelines to ensure that students learn with students of different racial and socioeconomic backgrounds (G. Orfield & Frankenberg, 2011). In 2007, the Supreme Court ruled that Louisville-Jefferson Countys assignment plan was unconstitutional, stipulating that districts could not use the individual race of a student as the sole determining factor in assigning students to schools (Parents Involved in Community Schools v. Seattle School District No. 1, 2007). In the aftermath of the Parents Involved decision, Louisville-area school officials worked with legal experts to craft a new multi-criteria assignment policy that takes into account several neighborhood-level demographic factors like family income, parent education level, and race (G. Orfield & Frankenberg, 2011).

Second behind Louisville on the spectrum is Chattanooga-Hamilton County, where a 1997 district merger spurred renewed interest in magnet schools. However, certain policy changes, like moving from race-conscious admissions policies to race-neutral ones and from providing more wide-ranging transportation to more limited transport (Siegel-Hawley et al., 2011), may have unfortunately limited the desegregating potential of Chattanooga-Hamilton County's magnet schools. The programs remain the sole method of bringing students together across the merged, city-suburban school system.

Towards the other side of the continuum, since unitary status and the subsequent implementation of a new student assignment policy in 2002 (dubbed the Family Choice Plan), schools in Charlotte-Mecklenburg have resegregated at an alarming pace (Bhargava, Frankenberg, & Le, 2008; Mickelson, Smith, & Southworth, 2009; Smith 2004). Area desegregation experts described the Family Choice Plan as essentially a neighborhood school assignment plan with race-neutral choice among nearby magnet schools (Mickelson et al., 2009, p. 136). The plan helped re-connect patterns of residential segregation to school enrollments, given the heavy emphasis on providing families with the option of attending their neighborhood school.

Finally, at the far end of the contemporary policy spectrum lies the Richmond area, where the failure to consolidate Richmond, Henrico, and Chesterfield in 1973 cemented the segregating impact of district boundary lines (Pratt, 1991). No stated desegregation policy has been pursued in the Richmond area since the 1980s (Ryan, 2010). Within the city limits, the basic principle of neighborhood schools remains somewhat complicated by the districts open enrollment policy. Until 2007, free transportation was provided to all students enrolling in elementary schools within their megazone (Richmond City is divided into three large zones, and students in those areas are allowed to attend any elementary school within their broad zone provided space is available once neighborhood students are enrolled). When transportation was cut backlikely the result of the recent fiscal climatecontroversy erupted. Some stakeholders, including the head of the Richmond Parent-Teacher Association, suggested that the open enrollment policy unfairly benefitted White and middle class parents in the city who could more easily find the means to provide transportation (School zones rekindling segregation, 2008).

Despite the varied policy contexts of the districts today, each of the metros experienced a basic post-Brown pattern in years past: token compliance with the decision, followed by a shift towards comprehensive desegregation efforts after the 1968 Green ruling outlawed freedom-of-choice plans. The advent of the 1970s marked the real point of departure, as Charlotte-Mecklenburg County and Louisville-Jefferson County began integrating schools across city-suburban boundary lines (Phillips, Rodosky, Munoz, & Larsen 2009; Smith, 2004), in addition to seeking to combine school desegregation efforts with housing desegregation policy (G. Orfield, 1981).

In fact, Louisville-Jefferson County was one of the only districts nationwide to include exemptions for integrated neighborhoods in its school desegregation plan (G. Orfield, 1981). The exemptions fell under three categories. First, neighborhoods meeting racial balance goals established in the school desegregation court order were excused from the school desegregation plan. Second, Black families making an integrative move into a predominately White neighborhood through the use of housing vouchers were exempted from busing. Third, neighborhoods that evolved into integrated environments were released (G. Orfield, 1981). In this manner, residential areas in the metro Louisville area were incentivized to become more diverse. And they did: Research on the time period from 1970 to 1980 indicates that school desegregation helped spur the first decline in housing segregation in three decades (Briley, 1985). Yet despite these positive early developments, Louisville-Jefferson Countys exemption policies overlooked the critical process of public housing site selection (G. Orfield, 1981). New public housing could be located in historically segregated neighborhoods, throwing off the racial balance component of the school desegregation policy.

Like metro Louisville, Charlotte blazed an early path in efforts to link school and housing desegregation policy (Smith, 2004). The contours of Charlottes strategy differed widely from Louisvilles, however, focusing on the location of public housing instead of the distribution of vouchers or certificates. A 1973 lawsuit detailing evidence that public housing had been intentionally sited in racially isolated neighborhoods eventually led to a settlement decreeing that approximately 900 scattered-site housing units had to be built across the metro area (G. Orfield, 1981; Smith, 2004). Importantly, school officials were consulted during the site planning process, leading one researcher to note, Charlotte was the only jurisdiction I visited [of twelve metros identified as having some model of school and housing coordination] where both planners and school officials mentioned that school integration considerations were a significant concern in the citys planning process (G. Orfield, 1981, p. 67). Unfortunately, those initial efforts remain the most noteworthy to date. In more recent years, the political interests of developmentand the associated windfall profitstook precedence over equity.

Meanwhile, in the 1970s-era Richmond and Chattanooga areas, school desegregation was limited to the central cities and efforts to join school and housing desegregation efforts were not attempted in any significant way.

The 80s and 90s witnessed a shift towards voluntary school integration for all of the metros, with each reporting efforts to create and expand magnet programs. These developments mirrored trends in districts across the nation and were incentivized by federal policy and an increased emphasis on school choice (Siegel-Hawley et al., 2008; G. Orfield, 1978). Richmond Citys efforts eventually fell far short of the establishment of a broad magnet system, whereas Chattanoogas early magnet implementation blossomed into an expanded set of programs after the 1997 merger (Pratt, 1990; Keller, 2006).

In the Charlotte and Louisville areas, magnet schools were established in the 1990s alongside system-wide student assignment plans that sought to promote diverse schools. Louisville-Jefferson County moved to implement an assignment policy based on controlled choice (Phillips et al., 2009). Controlled choice allowed parents some latitude in student placement, though the district made final assignment decisions in service of creating racially diverse academic settings. Unlike Louisvilles largely successful transition to a choice-based desegregation plan, however, in Charlotte the expanded emphasis on magnet programs in the early 90s signaled the beginning of a slow unraveling of what had been the first completely desegregated school system in the country (Bhargava et al., 2008). Visionary proposals to combat Charlottes housing segregation were also subverted during the 1990s (Smith, 2004).  

At present, three of the four sites continue to pursue some manner of school desegregation strategy. None of the metros operates under housing policy designed specifically to promote integration (as neither Charlotte nor Louisville continued earlier efforts to join school and housing policy). Though only one district, Louisville-Jefferson County, still implements comprehensive city-suburban school desegregation, Charlotte-Mecklenburg County employed similar prior practices, with policy shifts occurring within the time period under study. The Charlotte area thus offers an important before and after portrait of a site that engaged with and then retreated from the goals of Brown v. Board of Education (1954). Chattanooga-Hamilton County also provides an example of a district experiencing a significant policy change between 1990 and 2010, with the school-district merger occurring in 1997. The Richmond area follows the traditional northern and western model of central city desegregation absent suburban involvement. As such, the four sites offer a critical comparative perspective on the relationship between boundary lines, desegregation policy, and racial segregation patterns in the metropolitan South.

DATA AND METHODS

School-level enrollment and segregation trends were derived from the National Center for Educational Statistics Common Core of Data (NCES CCD). NCES is a reliable data source that collects federal school enrollment figures from virtually every district in the nation (Frankenberg, Lee & Orfield, 2003). NCES CCD provided racial/ethnic and free and reduced price lunch (FRL) enrollment data for the school years 1992 (the earliest year numbers are available at the school level), 1999-2000, and 2008-09.

In terms of residential segregation, block group-level information from the U.S. Census Bureaus decennial census Summary File 3 (SF3) helped describe the distribution and segregation of persons by race for the years 1990 and 2000. Summary File 3 data provides more detailed population and housing information from a large sample of the universal census count (U.S. Census, 2010). Redistricting data by race were used for 2010, as Summary File 3 was not yet available. Released first for immediate use in redrawing political boundaries every decade, the redistricting data provide full population counts by race for all levels of geography. The redistricting data is less comprehensive than census information released later pertaining to age, sex, households, families, and housing units but is highly reliable and useful for research dedicated to tracking racial trends.

Block groups were chosen as the unit of analysis because recent research indicates that block group-level data most accurately depicts trends in residential segregation (Bischoff, 2008; Mitchell, Batie, & Mitchell, 2010; Wilson, 2011). Block groups also correspond most directly with our understanding of neighborhoods, unlike the much larger census tracts (Grannis, 2005; Steinmetz & Iceland, 2003).

This analysis employed sample counts of the entire populationnot simply the school-aged populationin an effort to better understand how comprehensive school desegregation policy might impact housing trends over different generations. One component of the theoretical foundation for the study suggests that school desegregation has perpetuating effects (Wells & Crain, 1994), in part meaning that students growing up under a school desegregation policy are more likely to seek out integrated neighborhoods later in life. Examining the wider population counts is thus in keeping with part of the theory guiding this analysis.

A note also about the use of metro area or metro or area to describe the four sites under study: here the terms simply refer to the geographic area encompassed by the school-district merger (or proposed merger, as was the case for the Richmond metro). This geographic unit is much smaller than a census-defined Metropolitan Statistical Area (MSA). But since the metro boundaries defined by this analysis encompass only areas directly affected by the school desegregation plan, they provide the most pertinent school and housing data.

The mapping component of the project drew upon Census TIGER/line Shapefiles, a geographic data format that provides spatial information for use in mapping software. Housing population trends were mapped using Census TIGER/line Shapefiles from the 1990, 2000, and 2010 U.S. Census counts.

A variety of primary and secondary sources provided information about the evolution of desegregation policy in each of the four sites. These sources included court documents, newspaper articles, research publications, books, and several informal interviews conducted via email and phone.

MEASURES OF SEGREGATION AND ANALYTIC STRATEGY

The dissimilarity index (D) is the most well-known and popular measure of segregation (Massey & Denton, 1988; Reardon & Firebaugh, 2002). The index reveals the proportion of persons in a particular racial/ethnic group that would have to move in order to achieve an even spatial distribution of the races (Duncan & Duncan, 1955; Massey & Denton, 1988, 1993). Interpreted on a 0-100 scale, larger numbers indicate higher levels of segregation for minority groups, so that 100 would mean perfect segregation, while 0 implies perfect integration. For instance, if Baltimore, Maryland had a hypothetical Black-White D of .86, 86% of Black residents in the city would have to move in order to achieve perfect integration with the White population.

The analysis of segregation trends was limited to White and Black populations for the purposes of this article, due in part to the still small proportions of Latino, Asian, and multiracial students and residents in the four sites (though Latino students make up an increasingly significant share of the population and should certainly be more closely examined in future research). The Souths painful legacy of slavery and Jim Crowas well as the breadth and scale of school desegregation efforts in the regionalso made it particularly critical to understand how this history has shaped existing patterns of segregation for Blacks and Whites.

Separate sets of maps, for each of the three different points in time, were constructed to describe segregation patterns in the four metro areas. Decennial census data for 1990, 2000, and 2010 were joined with Census TIGER/line Shapefiles to produce GIS maps showing housing patterns at the block-group level by race. Elementary school addresses in the four metros were also geocoded into the mapping software, allowing a pie graph illustrating the racial and socio-economic enrollment for each school to appear on the actual school site.

Though overall segregation calculations were based on all regular public schools in the districts, the maps relied upon elementary school data. Because elementary schools are associated with the smallest attendance zones or catchment areas, their enrollment patterns most closely reflect underlying housing patterns in districts without ameliorating desegregation policy (M. Orfield, 2002; M. Orfield & Luce, 2010). The visual component of this study was thus aided by elementary school trends that best indicate whether school and housing patterns are closely aligned or whether school desegregation policies that deemphasize boundary lines help break the link between the two.

ANALYSIS

This section underscores two critical findings: 1) Metro areas implementing school desegregation strategies that bridge city-suburban boundary lines reported significantly faster rates of decline in housing segregation over time and 2) Metro areas implementing comprehensive school desegregation plans decoupled patterns of school segregation from housing segregation in a striking manner, so that even when residential isolation remained high, students experienced integrated school settings. Moreover, when desegregation plans were dismantled, patterns of school segregation quickly rose to meet and mirror underlying residential segregation trends.

To illustrate these findings, the first half of this section presents elementary school enrollment by race and poverty and housing patterns in tandem with one another for Black residents (beginning with a set of maps to visually highlight the relationship). The second half provides an analysis of Black-White school and residential segregation trends using the index of dissimilarity.

 

MAPPING SCHOOL AND HOUSING SEGREGATION TOGETHER

Spatial images of the distribution of the Black population of the four metros, overlaid with the racial composition of metro elementary schools, provide a telling portrait of the way school desegregation policy can disrupt underlying housing patterns. Figures 2-5 show that, while each of the four southern metros contained areas of intense racial isolation for Black residents (the deepest shade represents census block groups where the Black population exceeds 90%), places with city-suburban school desegregation policies were able to distribute students of different racial backgrounds much more evenly across the metropolitan landscape.

In the early 1990s, the contrasts between Louisville-Jefferson County and Charlotte-Mecklenburg County and the Richmond and Chattanooga areas were arresting (see Figure 2). The former two metros displayed trends showing a strong disparity between patterns of residential and school segregation. Note, for example, the presence of racially balanced schools in Charlotte-Mecklenburg and Louisville-Jefferson County even in highly segregated neighborhoods where Black residents made up 90% or more of the population. The other two metrosRichmond and Chattanooga (pre-merger)were defined by patterns showing that Black students attended racially isolated elementary schools that were, in turn, surrounded by intensely segregated Black neighborhoods. During the early 1990s, the vast majority of these segregated schools and neighborhoods were located within the school-district boundaries (illustrated by heavy black lines on the maps) encircling Richmond and Chattanoogas urban cores.

In the late 90s and through the 2000s, the Richmond and Chattanooga areas continued to report elementary school enrollments that largely mirrored nearby block groups. As Richmonds suburbs became more diverse, so too did their school enrollmentsbut the predominant pattern of racial isolation in the schools and neighborhoods of the inner suburbs and urban core remained consistent. Chattanooga-Hamilton County reported similar spatial patterns, suggesting that the expanded magnet school policy implemented at the time of the 1997 merger may not have been effectively altering the link between school segregation and underlying residential patterns.

Meanwhile in North Carolina, Charlotte-Mecklenburgs earlier success in disrupting the relationship between school and housing segregation gave way in the face of mounting political opposition and the 1999 unitary status decision. By 2000, schools in and around Charlottes urban core (particularly in the west and northwest quandrants of the city) were beginning to reflect the underlying Black population of area neighborhoods. Ten years later, Charlotte-Mecklenburg school enrollments had become extremely linked to patterns of neighborhood racial isolationan unsuprising fact, given the post-unitary status emphasis on neighborhood schools (see Figure 4). The early disconnect between patterns of residential and school segregation had all but disappeared.

With these developments unfolding in Charlotte-Mecklenburg schools, Louisville-Jefferson County became the only metro under study to consistently disrupt the relationship between Black-White school and housing segregation. Figures 2-4 display the impact of the metros ongoing commitment to a comprehensive city-suburban school desgregation policy. It should be noted, though, that by 2008, several elementary schools in Louisvilles still highly segregated inner city had become predominately Black and Latino. This uptick in isolation may have been related to the struggle to craft and implement a revised student assignment plan that would comply with the 2007 Parents Involved decisoin (the new plan was formally implemented in September 2009, and more recent changes were approved in January 2012).

Interestingly, despite the varied sucessses of the racial desegregation plans, patterns from each of the four metros indicated that racial isolation for Black residents was linked to elementary schools with high levels of student poverty. Using housing data from 2010 and school enrollment data from 2008, Figure 5 shows that nearly all schools in neighborhoods with high concentrations of black residents contained a majority of students qualifying for free or reduced priced lunches (signified by the deep green shade on the pie charts; non-poor students are represented with blue). Conversely, block groups with very low counts of Black residents were linked to schools with much larger shares of non-poor students. These broad trends, which persisted even in the metro with an on-going, wide-ranging school desegregation plan, could reflect the historic emphasis on integrating students by race, but not necessarily by class. Importantly, Louisville-Jefferson Countys multicriteria student assignment planwhich considers socieoconomic status alongside racemay produce a different set of patterns if it is fully implemented in coming years.

Figure 2. Elementary school racial composition by percent Black population living in census block groups, four metropolitan areas, 1990 and 1992-1993

Source: NCES Common Core of Data, 1992-93. U.S. Census 1990, SF3, P012.

Figure 3. Elementary school racial composition by percent Black population living in census block groups, four metropolitan areas, 2000 and 1999-2000.

Source: NCES Common Core of Data, 1999-2000. U.S. Census 2000, SF3, P007.

Figure 4. Elementary school racial composition by percent Black population living in census block groups, four metropolitan areas, 2010 and 2008-09.

Source: NCES Common Core of Data, 2008-09. U.S. Census 2010, P02, redistricting data.

Figure 5. Elementary school poverty composition by percent Black population living in census block groups, four metropolitan areas, 2010 and 2008-09.

Source: NCES Common Core of Data, 2008-09. U.S. Census 2010, P02, redistricting data.

LOW LEVELS OF SCHOOL SEGREGATION LINKED TO SIGNIFICANTLY FASTER DECLINES IN HOUSING SEGREGATION

In the early 1990s, at a time when both the Louisville and Charlotte metros were implementing far-reaching, city-suburban desegregation plans, the disconnect between patterns of school and housing segregation in the two metros was clearly evident (see Figure 6). Using the most widespread measure of segregationthe dissimilarity indexto broadly describe the distribution of Blacks and Whites across the metro, Louisville-Jefferson County and Charlotte-Mecklenburg both reported extremely low levels of Black-White school segregation. About 20-25% of Black students in both metros would have needed to change schools to be perfectly integrated with Whites. At the same time, the Louisville and Charlotte areas displayed much higher levels of housing segregationbetween 70% and 80% of Black residents needed to move to achieve perfect integration with Whites, or vice versa. During the same time period, the two metros lacking policy that sought to desegregate students or bridge the urban-suburban boundary linethe Richmond and Chattanooga areasreported high levels of both school and housing segregation. These trends provide support for the power of metropolitan school desegregation and the perpetuating effects of integrated schooling experiences, the theoretical elements undergirding the analysis.

Between 1990 and 2000, while Black-White residential segregation declined in all four sites, the two metros with large-scale, city-suburban desegregation plans reported swifter decreases in housing segregation levels (see Figures 7 and 8). For example, in the Charlotte and Louisville areas, the dissimilarity index indicated that Black-White housing segregation fell by approximately 10%, compared to a roughly 5% decline in the Richmond area and an 8% reduction in Chattanooga-Hamilton County (a figure that may be related to the 1997 merger). This trend is consistent with prior research (Frankenberg, 2005; G. Orfield, 2001; Parents Involved v. Seattle, 2006) and suggests that school desegregation policy may have influenced housing patterns, though, of course, this methodology cannot prove a causal relationship.

Figure 6. School and residential Black-White/White-Black dissimilarity index, four metropolitan areas, 1990-2010.

    

Source: NCES Common Core of Data, 1992-1993, 1999-2000, 2008-2009; U.S. Census 1990, 2000, 2010.

Figure 7. Percent change in Black-White/White-Black residential dissimilarity index, four metropolitan areas, 1990-2000, 2000-2010 and 1990-2010.

Source: NCES Common Core of Data, 1992-1993, 1999-2000, 2008-2009; U.S. Census 1990, 2000, 2010.

Later evidence from Charlotte-Mecklenburg (over the course of the first decade of the 2000s) crystallized the impact of ending a comprehensive school desegregation policy. The resegregation of Charlotte-Mecklenburg schools following the unitary status ruling quickly led to a situation where the relationship between school and housing segregation trends looked identical to that of the Richmond metro, where school and residential segregation patterns had long been virtually indistinguishable (Figure 6). Furthermore, between 2000 and 2010, Charlottes housing segregation began to decline at a much slower rate than it had during the previous decade, when the school desegregation policy remained in effect. The Black-White index of dissimilarity fell by about 5% between 2000 and 2010, compared to just over 10% between 1990 and 2000 (see Figure 7). Charlotte-Mecklenburg County went from leading the group of four metros in progress on Black-White housing segregation during the 1990s to third from last, just in front of the Richmond area, between 2000 and 2010.

The Richmond areadivided by city-suburban district boundaries and where no desegregation policy has been enacted since the mid 80sexperienced the slowest decline in Black-White school and housing segregation. Overall, Black-White residential dissimilarity declined about 12% in Richmond from 1990 to 2010, a much slower rate than Louisvilles and moderately slower than similar figures in the Charlotte and Chattanooga areas. With no comprehensive efforts guiding desegregation in Richmonds schools and neighborhoods, these are not necessarily surprising results. The two-decade overlap between school and housing segregation suggests that racial isolation in schools will continue to mirror persistently high patterns of neighborhood isolation in the Richmond area.

In Chattanooga-Hamilton County, a metro that experienced a significant policy shift just prior to 2000, dissimilarity trends for schools and neighborhoods showed that both housing and school segregation began to decline much more quickly following the 1997 merger. Between 1990 and 2000, Black-White school segregation only fell by about 4%, and housing segregation levels declined by just 8%. By contrast, from 2000 to 2010, when patterns associated with the late 1990s merger were more likely to appear, the dissimilarity index for Black-White school segregation dropped by about 20%, and housing segregation levels fell by almost 11% (Figure 1A in Appendix). However, the still largely intertwined relationship between school and housing segregation (see Figure 6) suggest that the system of magnet schools in place in the metro was not disentangling patterns of school segregation from housing trends in the same manner as the student assignment plans in the Louisville and pre-unitary Charlotte areas.

Finally, Louisville-Jefferson County provided an example of a metro with sustained, low levels of Black-White school segregation. In conjunction with that trend, Louisville-Jefferson County reported the fastest decreases in housing segregation. Black-White housing segregation in Louisville-Jefferson County fell roughly 9 percentage points from 1990-2000 and about 13 percentage points from 2000 to 2010. The faster rate of decline from 2000-2010 came in the midst of slightly rising levels of school segregation but may reflect other factors, including the 2000 municipal merger between the city and county. Findings from the Louisville area, located at one end of the spectrum of policy and boundary-line interventions to counter segregation, contrasted sharply with figures from Richmond, the metro at the opposite end. Black-White housing segregation in Louisville-Jefferson County fell almost twice as quickly as residential segregation trends in the Richmond area (see Figure 7). In short, evidence from Louisville-Jefferson County underscores the positive relationship between metropolitan school desegregation plans and declining rates of Black-White residential segregation.

It should be noted that similar results were derived from another measure of segregation, upholding the patterns of isolation in metro schools and neighborhoods reported here using the index of dissimilarity (see Appendix, Figure 2A). Taken together, these findings help show that school policy can indeed also become housing policy. Black-White housing segregation declined considerably faster in metros with a stable, city-suburban school desegregation plan. School segregation trends remained disconnected from high levels of housing segregation in areas where desegregation efforts were maintained and adapted over time but quickly rose to meet the intensity of housing segregation once school desegregation policy was dismantled.

DISCUSSION

This analysis presented patterns of school and residential desegregation together in an effort to illuminate the relationship between the two. Perhaps most striking among the findings was the way that school desegregation plans encompassing the community writ large disrupted the traditional influence of residential patterns on school enrollments. In support of James Madisons early Federalist philosophy, areas with strong school desegregation policy across city-suburban lines were able to significantly hasten declines in levels of Black-White housing segregation.

That trend was evident in both Charlotte-Mecklenburg and Louisville-Jefferson County during the early years of this study. Later, the example of Charlotte illustrated the manner in which school segregation becomes reconnected to housing trends once a desegregation policy is abandoned. Indeed, the example of Charlotte highlights the influence of the current political and legal context on the perpetuating effects of school desegregation. Those effects may not be enough to counteract broader social forces that have, since the early 1980s, contributed to the rollback of judicial oversight of desegregation efforts, to a heightened emphasis on competition and test scores, to the conflation of race and property values, and to widening income inequality (Boger & Orfield, 2005; Freund, 2007; Petrovich & Wells, 2005; Reardon, 2011; Wells, Duran, & White, 2011). As individuals who graduated from desegregated educational settings struggle to navigate contexts where these trends compete with their own values and understanding of the benefits of diverse schools, the perpetuating effects of their experiences may weaken (Wells et al., 2011).

In Chattanooga-Hamilton County, despite progress in reducing Black-White segregation after the merger, an intertwined relationship persistedreflecting the difficulty in overriding patterns of pervasive residential segregation with a limited system of countywide magnet schools. The Richmond metro, by contrast, has done nothing to intervene in patterns of school or housing segregation and trends in the area reflected that inaction.

Broadly, the analysis indicated that three significant forces contribute to patterns of school and housing segregation. First, demographic changes and accompanying shifts in residential segregation help produce housing patterns that surround schools. Second, in metros where school-district boundary lines remain intact, they give structure to regional patterns of school and housing segregation. Conversely, when those boundaries are subverted, patterns become less distinct. And third, the presence of school desegregation policy can significantly alter the contours of both school and housing segregation trends. This analysis of four southern metro areas shows each of the latter two forces must be positively aligned in order to produce real gains in integration; subtract one of them and major setbacks occur. For the one metro that has done nothing over the past two decades to attempt to foster school integration, the absence of any type of policy has meant virtually no gains at all.

In a nation experiencing rapid population shifts, evidence from these four metros emphasizes the on-going and urgent need for policies designed to equalize educational and life opportunities. The articles central conclusion is simply this: policy matters. Without it, the status quo trends negative.

Some education stakeholders continue to advocate for strategies that do little-to-nothing to shift the current course of segregated schools and housing, dismissing policy efforts to alter it as social engineering (Armor, 1995). Such arguments ignore the role of government policy in fostering contemporary trends. For subsidized highways slicing through city neighborhoods to carry middle class families out to the suburbs, lending practices underwriting suburbanization, and the selection of public housing project sites, federal and state governments are responsible. For exclusionary zoning policy, unchecked and often unscrupulous development, and racially identifiable school attendance lines, local government agencies should be held accountable (Freund, 2007; Massey & Denton, 1993). Together with widely documented housing discrimination (Hartman & Squires, 2010; National Commission on Fair Housing and Equal Opportunity, 2008) and, in most regions, overwhelming inaction in taking affirmative steps to counter the fallout from these decisions, key actors at all levels of the public and private sector continue to be complicit in producing a modern-day version of Jim Crow.

These overarching circumstances, accompanied by the fundamental fact that nearly all segregated environments remain desperately unequal (Anyon, 2005; Grant, 2009; Ryan, 2010), mean we must push back hard against a line of reasoning suggesting we should be content with current realities. Growing shares of under-served, non-White, and poor schoolchildren, along with their White and non-poor counterparts who will grow up in a multiracial society, depend upon policy actionrather than inactionto help shape a better, more viable future for coming generations.

LIMITATIONS

Though this study provided valuable new evidence affirming the importance of metropolitan school desegregation strategies, it contained several limitations. The absence of school and housing segregation data from the years prior to 1990 meant that the analysis failed to capture the effectiveness of metro desegregation policy during the years when it was most powerfully implemented. However, previous studies have documented the desegregation of schools and associated academic benefits in Charlotte-Mecklenburg and Louisville-Jefferson County (Kurlaender & Yun, 2002; Parents Involved v. Seattle, 2006; Phillips et al., 2009; Smith, 2004), and this research sought to extend upon prior work to understand more recent trends.

This study also limited the examination of school and housing segregation trends to the metro area encircled by the consolidated school-district boundary lines (or, in the case of Richmond, the area of the metro that would have been consolidated had the lower court ruling stood). This methodological decision allowed for the most precise understanding of how school desegregation policy impacted housing segregation, since the areas under study were directly under the influence of the school policy. Yet three of the four merger decisions occurred in the 1960s and 70s, and much outward suburban growth and development has occurred in the intervening years. As a result, this analysis did not capture a larger portrait of metropolitan change, limited as it was to the city and suburban jurisdictions involved in the desegregation plan (or, again, in the case of Richmond, the proposed plan).

Though certain aspects of this study provided insight into the impact of school choice on segregation patterns (e.g. the open enrollment policy in the Richmond metro or the magnet school program in Chattanooga-Hamilton County), it did not delve specifically into the way different types of choice affected segregation levels. The historical presumption that students should attend schools closest to homereinforcing the school-housing segregation relationshiphas increasingly been called into question by the growth and popularity of school choice. Charter schools, for example, have become a dominant presence in Charlotte-Mecklenburg, where a system of magnet schools and a controlled choice policy are also present. It would be important to understand how each of these newer manifestations of choice is linked to school and residential segregation.  

Lastly, the four sites under study provided critical insight into the effectiveness of metropolitan school desegregation policy in the region of the country where those strategies were most fully pursued. The diverse demographic and policy contexts of each site promoted the generalizability of the findings, but the attention to the southern region of the U.S. may limit more national implications. Still, the emphasis on the South was somewhat mitigated, because the Richmond metros policy context is not unlike more fragmented metro areas in other parts of the country.

DIRECTIONS FOR POLICY

This study provided evidence to support the still nascent movement towards more regional solutions for metropolitan inequalities. The regional agenda underlines the importance of metropolitan elasticity, or the ability to minimize harmful development and sprawl on the outer edges of a metro through a number of different avenues (Rusk, 1999). Each of these avenues involves efforts to subvert boundary lines, either through municipal consolidation or annexation processes or heightened regional cooperation.

The federal government should provide unequivocally strong leadership supportive of regional efforts to combat metro area inequality. The creation of a high-profile commission or task force to investigate innovative, region-wide school and housing integration strategies would be a positive first step. Federal funding for a regional schools and housing initiative would also signal high-level leadership and interest in promoting more equitable metropolitan areas.

State legislatures should work to support metropolitan elasticity, either through the enactment of policy that stimulates municipal and school district consolidation or annexation or with incentives to promote regional strategies. Virginia, for example, passed a law in 1996 that provides money for municipalities that collaborate through regional partnerships. Regions are most heavily rewarded for plans to cooperate around education, revenue-sharing, human services, local land use, and housing (Rusk, 1999).

State law and policy permitting, the task of framing the political debate surrounding annexation or consolidation in terms that are agreeable to populations who have often been pitted against one another is immense but not insurmountable (Leland & Thurmaier, 2004). City-county consolidation efforts require that narrow notions of neighborhood and community be expanded to include a diverse cross-section of interests and needs. Referendum campaigns that explore the fundamental interests of the local community and devise their political strategy accordingly have a higher probability of winning public support. Strong local leadership and agenda-setting from business, media, and community stakeholders is also critical (Leland & Thurmaier, 2004).

Despite the political difficulty linked to mergers, in the current fiscal climate, conversations around school district consolidation have emerged with some frequency. Many recent consolidation efforts have been spurred by the potential financial benefits that may come with the elimination of duplicative bureaucracies, with some notable exceptions. The Memphis School Board, for example, recently voted to surrender the school systems charter and consolidate with Shelby County, its neighboring suburban district. Events in Tennessee are still in progress but bear close watching as an example of a contemporary effort to bridge district boundaries in the name of equity.

Of course, in the absence of efforts to actually eliminate boundary lines, other alternatives that ease the segregating effects of those lines should be considered. With respect to efforts to promote educational opportunity across regional boundaries, a number of options beyond actual consolidation are available. Pairing nearby city and suburban schools (or inner suburban and outer suburban schools), one with declining enrollment, the other with a burgeoning student population, seems like fiscal commonsense. Finding ways to leverage school choice across boundary lines is another consideration. Regional magnet or charter schools, imbued with appropriate civil rights protections, might offer integrative possibilities. The nations eight long-standing, inter-district transfer programs could also be expanded. Finally, the student transfer provision under ESEA, while currently underutilized, may deliver inter-district possibilities (Holme & Wells, 2009).

While the Milliken decision presents a difficult legal barrier to crossing school district lines, fair housing law, if proactively enforced, could advance regional housing opportunities (which, in turn, would positively influence school diversity efforts). The federal government should heighten its efforts to expand and implement the 1968 Fair Housing Act. In addition to fair housing enforcement possibilities, the federal government could work to broaden the Moving to Opportunity program to other cities. All levels of government may help ensure that Section 8 and the Low-Income Housing Tax Credit are disbursed in ways that promote affordable housing in high-opportunity areas throughout a region (M. Orfield & Luce, 2010; Pfeiffer, 2009). Law and policy should also guarantee that new metropolitan development contains a certain share of affordable housing (Rusk, 1999).  

Each of the policy options described above should be crafted with an eye towards jointly disrupting patterns of school and housing segregation. For much of the time period under study, school desegregation policy bore the full weight of responsibility for interrupting underlying patterns of residential isolation. Given the success of those one-dimensional efforts in the Louisville and pre-unitary Charlotte areas, it is difficult to comprehend the power of a joined school and housing desegregation strategy. Early, pre-1990s evidence from the same two metros provided extremely rare examples of coordinated school and housing desegregation efforts (G. Orfield, 1981). The programs did not continue but offer ideas for how regional communities might begin to design united housing and school policy. For instance, voluntary school integration programs should consider transportation exemptions for families making integrative moves, in addition to providing exceptions for residents of stable, diverse communities. There is also a need for expertise in the areas of housing and schools to flow across metropolitan districts and agencies. Local housing programs develop and shiftas do student assignment plans and building and redistricting decisionswith little knowledge or discussion about the two related processes within the different sectors (G. Orfield, 1981; Tegeler, 2011). Public officials should be put into place who can help bridge these gaps.

CONCLUSION

Over the past half century, rapid growth and demographic change in metropolitan areas has been accompanied by a dearth of policy seeking to harness the potential of those transformations. Instead, in many ways, law and policy has cemented tremendous racial and socioeconomic inequities into the structure of our cities.

This analysis provides evidence of the consequences of our retrenchment of strategies designed to equalize educational opportunities. It revisited the region of the country most impacted by Browns lofty mandate, seeking to understand more fully the long-term effects of school desegregation policies coupled with attempts to render district boundary lines irrelevant.

In places where desegregation plans have been significantly altered, or dropped altogether, the findings tell a story of what was lostbut also what might still be regained. Areas that continue to operate under comprehensive metropolitan school desegregation policies are decidedly more likely to display integrated schools and residential communities. Other locales that have not yet experimented with policies designed to lessen the impact of boundaries and promote desegregation must examine the consequences of their inaction and be spurred towards greater efforts.

The underlying story, highlighted by the unsteady trajectory of school desegregation history, is that policy, law, and politics can and do change. While circumstances in some of the metro areas under study may have undermined progress towards more equal schools and neighborhoods, there is always the promise that subsequent policies will be different. The results of this analysis should inform those future efforts to promote vibrant and healthy regional spaces across the country.

Notes

1. Showing intentional discrimination is difficult for several reasons. After Brown, governmental units were aware that segregating policies were illegal and thus more careful not to create documents that would serve as evidence against them. Without direct evidentiary material, plaintiffs have to collect a vast amount of materials showing the history of school construction, rezoning, and feeder patterns and decisions about the formation of boundary lines, among other things. Building such a case was typically beyond the reach of civil rights groups and lawyers, already pressed for time and resources (G. Orfield, 1978).

2. Largely due to the increasing impermeability of district and municipal boundary lines, the country now struggles with a situation where a vast proportion of both school and housing segregation can be attributed to racial separation between districts or municipalities, rather than within them (Bischoff, 2008; Clotfelter, 2004; Farrell, 2008; Reardon & Yun, 2002; Weiher, 1991).

3. The Louisville-Jefferson County School Board recently approved a revised student assignment plan by a unanimous vote. The new plan would allow many students to attend schools closer to their homes, while still maintaining a focus on promoting racial diversity. Language was added to the diversity index for the first time (Konz, 2012).

4. Rather than ask families to submit annual income reports, most American school systems use the percent of children qualifying for free and reduced lunch as a measure of poverty at the school and district level. In order for students to be considered eligible for free and reduced lunch prices, family income must fall within a set of guidelines established yearly by the United States Department of Agriculture (USDA). For 2007-08, the federal poverty level for a family of three was $17,170. To qualify for reduced priced meals, the same family of three would have to fall within 185% of the poverty level, with an annual income at or below $31,765. To receive free meals, annual income had to be within 130% of the federal poverty level, or $26,845. These figures are significant, as they demonstrate both the low estimates of poverty in general, and for children in particular. This study thus uses the terms FRL-eligible, low-income, and poor interchangeably. Though research has shown eligibility for free and reduced priced lunch (a popular measure of relative student poverty) to be somewhat problematic (Harwell & LeBeau, 2010), it is still widely used due to the easy availability of the data.

5. The one exception for the school year 1999-2000 is Chattanooga-Hamilton County, where data were not available. NCES data from 1998-1999 was used instead.

6. American Indian students were excluded from the analysis for all school years due to the small size of the population in the four metros studied (less than 1%). Students identifying as two or more races or Hawaiian for the 2008-09 school yearthe first year these data were collectedwere excluded for the same reason, as they represented less than 1% of the population in each of the metros.

7. It should be noted here that dealing with the reorganization of census boundary lines is an important issue when examining changes in populations over time. A drawback to using block groups instead of tracts for this analysis was that relationship/comparability files that help standardize longitudinal boundary shifts were unavailable (Du, Coles, OCampo, & McNutt, 2007). Given the benefits of block group level data discussed in the text, however, the tradeoff was considered worthwhile. Importantly, tract-level figures derived from one of the nations leading centers for U.S. Census 2010 research, in U.S. 2010: America in the First Decade of the New Century, provide support for residential segregation trends in the four metros that were reported in this article (see http://www.s4.brown.edu/us2010/Data/Data.htm). Also, since most population-related census boundary changes divide larger areas into smaller units of geography, a logical extension of the empirical rationale for using the lowest level of census data possible (Bischoff, 2008; Mitchell, Batie, & Mitchell, 2010; Wilson, 2011) would suggest the patterns reported here may actually underestimate the level of segregation present in the four metros.

8. In a few instances, census block group data for the 1990 and 2000 decennial counts were missing. In the maps, if tract-level data for the same year was available, it was layered under block group data to fill in the gaps. See Table 1A in the Appendix for a list of missing block groups by tract number in the metro areas under study.

9. For example, the Charlotte MSA includes five North Carolina counties and one county in South Carolina. For the purposes of this study, though, the Charlotte area refers to Charlotte City and Mecklenburg County (which encircles the city).

10. Richmond, of course, is a special case, since the metro area did not experience consolidation. This analysis defines the Richmond metro as the city and two surrounding counties that were targeted under the original consolidation proposal.

11. In addition, the vast majority of school desegregation plans in the region emphasized the desegregation of Black students and Whites (G. Orfield, 1978, 2007).

12. The two different sets of years noted in the title of the earliest school-housing figures indicate that the decennial census count was used to illustrate or calculate patterns of residential segregation in 1990, while school trends were derived from the 1992-93 school year, the earliest for which data were available at the school level for all four metros in NCES CCD.

13. Some predominately Black suburban communities, particularly in the Richmond area, are linked to schools with higher shares of non-poor students.

14. The slow decline in Black-White housing segregation in the four metros mirrors national trends over the same time period (Frey, 2010).

15. Recent research shows that district consolidation does not always result in financial gains (Howly, Johnson, & Petrie, 2010).

References

Anyon, J. (2005). Radical possibilities: Public policy, urban education and a new social movement. New York, NY: Routledge.

Armor, D. (1995). Forced justice: School desegregation and the law. New York, NY: Oxford University Press.

Balfanz, R., & Legters, N. E. (2008). NCLB and reforming the nation's lowest-performing high schools. In G. L. Sunderman (Ed.), Holding NCLB accountable: Achieving accountability, equity, and school reform (pp. 191-207). Thousand Oaks, CA: Corwin Press.

Beck, J. H. (1983). Tax competition, uniform assessment, and the Benet principle. Journal of Urban Economics, 13, 127-146.

Bhargava, A., Frankenberg, E., & Le, C. (2008). Still looking to the future: Voluntary K-12 school integration. New York, NY: NAACP Legal and Defense Fund/Civil Rights Project at UCLA.

Bischoff, K. (2008). School district fragmentation and racial residential segregation: How do boundaries matter? Urban Affairs Review, 44(2), 182-217.

Boger, J., & Orfield, G. (Eds.). (2005). School resegregation: Must the South turn back? Chapel Hill, NC: University of North Carolina Press.

Braddock, J. (1980). The perpetuation of segregation across levels of education: A behavioral assessment of the contact-hypothesis. Sociology of Education, 53, 187-186.

Braddock, J. (2009). Looking back: The effects of court-ordered desegregation. In C. Smrekar & E. Goldring (Eds.), From the courtroom to the classroom: The shifting landscape of school desegregation (pp. 3-18). Cambridge, MA: Harvard Education Press.

Bradley v. Richmond School Board, 416 U.S. 696. (1974).

Braddock, J. (1980). The perpetuation of segregation across levels of education: A behavioral assessment of the contact-hypothesis. Sociology of Education, 53, 187-186.

Braddock, J. (2009). Looking back: The effects of court-ordered desegregation. In C. Smrekar, C. & E. Goldring (Eds.). From the courtroom to the classroom: The shifting landscape of school desegregation (pp. 3-18). Cambridge, MA: Harvard Education Press.

Brief of Housing Scholars and Research & Advocacy Organizations as Amici Curiae in Support of Respondents, Parents Involved in Community Schools, Petitioner v. Seattle School District No. 1, et al.; Meredith v. Jefferson County Board of Education, U.S 551 (2006).

Brief of The Swann Fellowship, Former School Board Members, Parents and Children from the Charlotte Mecklenburg Schools as Amici Curiae in Support of Respondents, Parents Involved in Community Schools, Petitioner v. Seattle School District No. 1, et al.; Meredith v. Jefferson County Board of Education, U.S. 551 (2006).

Briggs, X. (2005). The geography of opportunity: Race and housing choice in metropolitan America. Washington DC: Brookings Institution Press.

Briley, K. (1985). School desegregation spurs first housing desegregation gain in forty years in Louisville and Jefferson County. Three decades of increased segregation erased between 1970 and 1980 (Staff Report 85-8). Kentucky Commission on Civil Rights.

Brown v. Board of Education, 349 U.S. 294. (1954).

Butler, D. (2010). Ethnic racial composition and college preference: Revisiting the perpetuation of segregation hypothesis. Annals of the American Academy of Political and Social Science, 26.

Charles, C. (2003). The dynamics of racial residential segregation. Annual Review of Sociology, 29, 167-207.

Chiodo, A., Hernández-Murillo, R., & Owyang, M. (2010). Nonlinear effects of school quality on house prices. Federal Reserve Bank of St. Louis Review, 92(3), 185-204.

Clapp, J., Nanda, A., & Ross, Stephen, L., (2008). Which school attributes matter? The influence of school district performance and demographic composition on property values. Journal of Urban Economics, 63(2), 451-466.

Clotfelter, C. T. (2004). After Brown: The rise and retreat of school desegregation. Princeton, NJ: Princeton University Press.

Crain, R. L. (1970). School integration and occupational achievement of Negroes. American Journal of Sociology, 75, 593-606.

Dawkins, M. P. (1983). Black students occupational expectations: A national study of the impact of school desegregation. Urban Education, 18, 98-113.

Di, Z. & Liu, X. (2005). The importance of wealth and income in the transition to homeownership. Cambridge, MA: Joint Center for Housing Studies at Harvard University.

Dougherty, J., Harrelson, J., Maloney, L., Murphy, D., Smith, R., Snow, M., & Zannoni, D. (2009). School choice in suburbia: Test scores, race, and housing markets. American Journal of Education, 115(4), 523-548.

Du, P., Coles, B., OCampo, P., & McNutt, L. (2007). Changes in population characteristics and their implication on public health research. Epidemiologic Perspectives & Innovations, 4(6).

DuBois, W.E.B. (1903). The souls of black folk. Chicago, IL: A.C. McClurg & Co.

Duncan, O., & Duncan, B. (1955). A methodological analysis of segregation indexes. American Sociological Review, 20, 210-17.

Eaton, S. (2001). The other Boston busing story: Whats won and lost across the boundary line. New Haven, CT: Yale University Press.

Farrell, C. (2008). Bifurcation, fragmentation or integration? The racial and geographical structure of U.S. metropolitan segregation, 1990-2000. Urban Studies, 45, 467.

 

Frankenberg, E. (2005). The impact of school segregation on residential housing patterns: Mobile, Alabama and Charlotte, North Carolina. In J. Boger & G. Orfield, ( Eds.), School resegregation: Must the South turn back? (pp. 164-186). Chapel Hill, NC: University of North Carolina Press.

Frankenberg, E. (2009). Splintering school districts: Understanding the link between segregation and fragmentation. Law and Social Inquiry, 34(4), 869-909.

Frankenberg, E., Lee, C., & Orfield, G. (2003). A multiracial society with segregated schools: Are we losing the dream? Cambridge, MA: Civil Rights Project, Harvard University.

Author et al. (2008). The forgotten choice? Magnet schools in a changing landscape. Los Angeles, CA: Civil Rights Project/Proyecto Derechos Civiles.

Freund, D. (2007). Colored property: State policy and White racial politics in suburban America. Chicago, IL: University of Chicago Press.

Frey, W. (2001). Melting pot suburbs: A Census 2000 study of suburban diversity. Washington, DC: The Brookings Institution.

Frey, W. (2010). Census data: Blacks and Hispanics take different segregation paths (The State of Metropolitan America, No. 21.). Washington, DC: Brookings Institution.

National Commission on Fair Housing and Equal Opportunity. (2008). Future of fair housing: Report of the National Commission on Fair Housing and Equal Opportunity. Washington, DC: The Leadership Conference on Civil and Human Rights. Retrieved from http://www.civilrights.org/publications/reports/fairhousing/.

Grannis, R. (2005). T-Communities: Pedestrian street networks and residential segregation in Chicago, Los Angeles, and New York. City & Community 4(3), 295-322.

Grant, G. (2009). Hope and despair in the American city: Why there are no bad schools in Raleigh. Cambridge, MA: Harvard University Press.

Green v. County School Board of New Kent County, 391 U.S. 430. (1968).

Hartman, C. & Squires, G. (Eds.). (2010). The integration debate: Competing futures for American cities. New York, NY: Routledge.

Harwell, M. R., & LeBeau, B. (2010). Student eligibility for a free lunch as an SES measure in educational research. Educational Researcher, 39, 120-131.

Hawley, W. (2007). Designing schools that use student diversity to enhance learning of all students. In E. Frankenberg & G. Orfield (Eds.), Lessons in integration: Realizing the promise of racial diversity in American schools (pp. 31-56). Charlottesville, VA: University of Virginia Press.

Haynes, B. (2001). Red lines, black spaces: The politics of race and space in a black middle-class suburb. New Haven, CT: Yale University Press.

Holme, J.J. (2002). Buying homes, buying schools: School choice and the social construction of school quality. Harvard Educational Review, 72(2), 177-205.

Holme, J.J., & Wells, A.S. (2008). School choice beyond district borders: Lessons for the reauthorization of NCLB from interdistrict desegregation and open enrollment plans. In R. Kahlenberg (Ed.), Improving on No Child Left Behind: Getting education reform back on track (pp. 139-215). Washington, DC: The Century Foundation.

Howly, C., Johnson, J., & Petrie, J. (2011). Consolidation of schools and districts: What the research says and what it means. Boulder, CO: National Education Policy Center.

Iceland, J., & Steinmetz, E. (2003). Using census block groups instead of census tracts when examining residential housing patterns (U.S. Census Bureau working paper). U.S. Census Burea. Retrieved from http://www.census.gov/hhes/www/housing/housing_patterns/pdf/unit_of_analysis.pdf.

Jargowsky, P. (1997). Poverty and place: Ghettos, barrios, and the American city. New York, NY: Russell Sage Foundation.

Jargowsky, P., & El Komi, M. (2011). Before or after the bell? School context and neighborhood effects on student achievement. In H. Newburger, E. Birch, & S. Wachter (Eds.), Neighborhood and life chances: How place matters in modern America  (pp. 50-72). Philadelphia, PA: University of Pennsylvania Press.

Johnson, H. B. (2006). The American dream and the power of wealth: Choosing schools and inheriting inequality in the land of opportunity. New York, NY: Routledge.

Kain, J. (1992). The spatial mismatch hypothesis: Three decades later. Housing Policy Debate, 3(2).

Kane, T., Riegg, S., & Steger, D. (2010). School quality, neighborhoods and housing prices: Effects of free choice among public schools. The Review of Economic Studies, 7 (3), 1164-1191.

Katz, B. (2001). Reflections on regionalism. Washington, DC: Brookings Institution Press.

Keller, B. (2006). Charging the gap. Education Week. Retrieved from http://www.pefchattanooga.org/Portals/0/Benwood/edweek1mar06.pdf.

Konz, A. (2012, January 9). JCPS board OKs revised student-assignment plan:

2,000 pupils could stay closer to homes. Louisville Courier-Journal. Retrieved from http://www.courier-journal.com/article/20120109/NEWS01/301090062/JCPS-board-OKs-revised-student-assignment-plan?odyssey=nav|head.

Kozol, J. (1992). Savage inequalities: Children in Americas schools. New York, NY: Harper Perennial.

Kurlaender, M. & Yun, J. (2005). Fifty years after Brown: New evidence of the impact of school racial composition on student outcomes. International Journal of Education Policy Research and Practice, 6(1).

Lareau, A. (2003). Unequal childhoods: Race, class and family life. Berekely, CA: UC Press.

Leland, S. M., & Thumaier, K. (Eds.). (2004). Case studies of city-county consolidation: Reshaping the local government landscape. New York, NY: M.E. Sharpe Publishing.

Linn, R., & Welner, K. (Eds.). (2007). Race-conscious policies for assigning students to schools: Social science research and the Supreme Court cases. Washington, DC: National Academy of Education.

Madison, J. (1787, November 22). The same subject continued: The union as a safeguard against domestic faction and insurrection. New York Daily Advertiser. Retrieved from http://www.montpelier.org/explore/james_madison/media/pdf/jm_federalist_10.pdf

Masnick, G. (2001). Home ownership trends and racial inequality in the United States

In the 20th century (Working paper W01-4). Cambridge, MA: Joint Center for Housing Studies, Harvard University.

Massey, D. S., & Denton, N. A. (1988). The dimensions of residential segregation. Social Forces, 67, 281-315.

Massey, D. S., & Denton, N. A. (1993). American apartheid: Segregation and the making of the underclass. Cambridge, MA: Harvard University Press.

Massey, D., Charles, C., Lundy, G., & Fischer, M. (2003). The source of the river: Social origins of freshmen at Americas selective colleges and universities. Princeton, NJ: Princeton University Press.

McDonnell, B. (27 January 2010). GOP response to President Barack Obama's State of the Union. Richmond, Virginia.

Mickelson, R., Smith, S., & Southworth, S. (2009). Resegregation, achievement, and the chimera of choice in post-unitary Charlotte-Mecklenburg schools. In C. Smrekar & E. Goldring (Eds.), From the courtroom to the classroom: The shifting landscape of school desegregation. Cambridge, MA: Harvard Education Press.

Mickelson, R. (2011). Exploring the school-housing nexus: A synthesis of social science evidence. In P. Tegeler (Ed.), Finding common ground: Coordinating housing and education policy to promote integration (pp. 5-8). Washington, DC: Poverty and Race Research Action Council.

Mickelson, R., & Bottia, M. (2010). Integrated education and mathematics outcomes: A synthesis of social science research. North Carolina Law Review, 88, 993.

Mitchell, D., Batie, M., & Mitchell, R. (2010). The contributions of school desegregation to housing integration: Case studies in two large urban areas. Urban Education, 45, 166.

National Fair Housing Alliance. (2006). Unequal opportunityperpetuating housing segregation in America (Fair Housing Trends Report). Washington, DC: National Fair Housing Alliance. Retrieved from http://www.mvfairhousing.com/pdfs/2006%20Fair%20Housing%20Trends%20Report.PDF

Nondiscrimination and Equal Opportunity Under Executive Order 11063: Proposed Rules, 44 Fed. Reg. (1976).

Oliver, M., & Shapiro, T. (1995). Black wealth, White wealth: A new perspective on racial inequality. New York, NY: Routledge.

Orfield, G. (1978). Must we bus? Segregated schools and national policy. Washington, DC: Brookings Institution Press.

Orfield, G. (1981). Toward a strategy for urban integration: Lessons in school and

 housing policy from twelve cities. A Report to the Ford Foundation.

Orfield, G. (2001). Metropolitan school desegregation: Impacts on metropolitan society. In Powell, J. & Kay, J. (Ed.), In pursuit of a dream deferred: Linking housing and education policy. New York, NY: Peter Lang.

Orfield, G. (2009). Reviving the goal of an integrated society: A 21st century challenge. Los Angeles, CA: UCLA Civil Rights Project/Proyecto Derechos Civiles.

Orfield, G., & Eaton, S. (1996). Dismantling desegregation: The quiet reversal of Brown v. Board of Education. New York, NY: The New Press.

Orfield, G., & McArdle, N. (2006). The vicious cycle: Segregated housing, schools and intergenerational inequality. Cambridge, MA: Civil Rights Project, Harvard University.

Orfield, G., & Frankenberg, E. (2008). The last have become the first: Rural and small town America lead the way on desegregation. Los Angeles, CA: UCLA Civil Rights Project.

Orfield, G., Frankenberg, E., & Garces, L. M. (2008). Statement of American social scientists of research on school desegregation to the U.S. Supreme Court in Parents v. Seattle School District and Meredith v. Jefferson County. The Urban Review, 40(1).

Orfield, G., & Frankenberg, E. (2011). Experiencing integration in Louisville: How parents and students see the gains and challenges. Los Angeles, CA: Civil Rights Project at UCLA.

Orfield, M. (2002). American metropolitics: The new suburban reality. Washington, DC: Brookings Press.

Orfield, M. (2006). Minority suburbanization, stable integration, and economic opportunity in fifteen metropolitan regions. Minneapolis, MN: Institute on Race and Poverty.

Orfield, M., & Luce, T. (2010). Region: Planning the future of the twin cities. Minneapolis, MN: University of Minnesota Press.

Parents Involved in Community Schools v. Seattle School District No. 1, 551 U.S. 701. (2007).

Pearce, D. (1980). Breaking down the barriers: New evidence on the impact of metropolitan school desegregation on housing patterns. Washington, DC: National Institute of Education.

Petrovich, J., & Wells, A. (Eds.) (2005). Bringing equity back: Research from a new era in American educational policy. New York, NY: Teachers College Press.

Pettigrew, T., & Tropp, L. (2006). A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology, 90(5), 751.

Pfeiffer, D. (2009). The opportunity illusion: Subsidized housing and failing schools in California. Los Angeles, CA: Civil Rights Project at UCLE.

Phillips, K., Rodosky, R., Munoz, M., & Larsen, E. (2009). Integrated schools, integrated futures? A case study of school desegregation in Jefferson County, Kentucky. In C. Smrekar & E. Goldring (Eds.). From the courtroom to the classroom: The shifting landscape of school desegregation. Cambridge, MA: Harvard Education Press.

Powell, J. (2009). Reinterpreting metropolitan space as a strategy for social justice. In M. P. Pavel. (Ed.), Breakthrough communities: Sustainability and justice in the next American metropolis. Cambridge, MA: MIT Press.

Powell, J., & Kay, J. (Eds.) (2001). In pursuit of a dream deferred: Linking housing and education policy. New York, NY: Peter Lang.

Pratt, R.A. (1991). The color of their skin: Education and race in Richmond, Virginia, 1954-89. Charlottesville, VA: University of Virginia Press.

Ready, D., & Silander, M. (2011). School racial and ethnic composition and young childrens cognitive development: Isolating family, neighborhood and school influences. In E. Frankenberg & E. DeBray (Eds.), Integrating schools in a changing society: New policies and legal options for a multiracial generation (pp. 91-113). Chapel Hill, NC: University of North Carolina Press.

Reardon, S. (2011). The widening academic achievement gap between the rich and the poor: New evidence and possible explanations. In G. Duncan & R. Murname (Eds.), Whither opportunity? Rising inequality, schools and childrens life changes (pp. 91-116). New York, NY: Russell Sage Press.

Reardon, S., Yun, J., & Eitle, T. (2000). The changing structure of school segregation: Measurement and evidence of multi-racial metropolitan school segregation, 1989-1995. Demography, 37(3), 351-364.

Reardon, S., & Yun, J. (2002). Suburban racial change and suburban school segregation, 1987-95. Sociology of Education, 74(2), 79-101.

Reardon, S., & Firebaugh, G. (2002). Measures of multi-group segregation. Sociological Methodology, 32, 33-67.

Rothwell, J. (2012). Housing costs, zoning, and access to high-scoring schools. Washington, DC: Brookings Institution.

Rusk, D. (1999). Inside game, outside game: Winning strategies for saving urban America. Washington, DC: Brookings Press.

Ryan, J. (2010). Five miles away, a world apart: One city, two schools and the story of modern educational inequality. New York, NY: Oxford University Press.

Schofield, J. (1995). Review of research on school desegregation's impact on elementary and secondary school students. In J. A. Banks and C. A.M. Banks (Eds.), Handbook of multicultural education. New York, NY: Macmillan Publishing.

School zones rekindling segregation, PTA says. (2008, February 6). Style Weekly. Retrieved from http://www.styleweekly.com/richmond/school-zones-rekindling-segregation-pta-says/Content?oid=1380033

Shapiro, T. (2004). The hidden cost of being African American: How wealth perpetuates inequality. New York, NY: Oxford University Press.

Smith, S. (2004). Boom for whom? Education, desegregation and development in Charlotte. Albany, NY: State University of New York Press.

Smrekar, C., & Goldring, E. (2000). Magnet schools and the pursuit of racial balance. Education and Urban Society, 33(17).

Sohoni, D., & Saporito, S. (2009). Mapping school segregation: Using GIS to explore racial segregation between schools and their corresponding attendance areas. American Journal of Education, 115, 569.

Southern Education Foundation (2010). A new diverse majority: Students of color in the Souths public schools. Atlanta, GA: Southern Education Foundation.

Stearns, E. (2010). Long-term correlates of high school racial composition: Perpetuation theory reexamined. Teachers College Record, 112(6).

Swann v. Charlotte-Mecklenburg Board of Education, 402 U.S. 1. (1971).

Siegel-Hawley et al. (2011). Integrating suburban schools: How to benefit from growing diversity and avoid segregation. Los Angeles, CA: Civil Rights Project at UCLA.

Tegeler, P. (Ed.). Finding common ground: Coordinating housing and education policy to promote integration. Washington, DC: Poverty and Race Research Action Council.

Tiebout, C. (1956). A pure theory of local expenditures. The Journal of Political Economy, 64(5), 416-424.

Timeline: Desegregation in Jefferson County public schools. (2005, September 3). Louisville Courier-Journal. Retrieved from http://www.courier-journal.com/apps/pbcs.dll/article?AID=2005509040428

Turner, M., & Ross, S. (2003). Housing discrimination in metropolitan America: Findings from the latest national paired testing study (Working paper 2003-08). Storrs, CN: University of Connecticut, Department of Economics.

U.S. Commission on Civil Rights. (1977). Statement on metropolitan school desegregation. Washington, DC.: U.S. Commission on Civil Rights.

Weiher, G. (1992). The fractured metropolis: Political fragmentation and metropolitan segregation. Albany, N.Y.: State University of New York Press.

Wells, A. S., Baldridge, B., Duran, J., Lofton, R., Roda, A., Warner, M., White, T., & Grzesikowski, C. (2009). Why boundaries matter: A study of five separate and unequal Long Island school districts. New York, NY: Long Island Index.

Wells, A. S., & Crain R. L. (1994). Perpetuation theory and the long-term effects of school desegregation. Review of Educational Research, 6, 531-555.

Wells, A. S., Duran, J., & White, T. (2011). Southern graduates of school desegregation: A double consciousness of resegregation yet hope. In E. Frankenberg & E. DeBray (Eds.), Integrating schools in a changing society: New policies and legal options for a multiracial generation (pp. 114-130). Chapel Hill, NC: UNC Press.

Wilson, W. J. (1991). Public policy research and the truly disadvantaged. In C. Jenks & P. Peterson (Eds). The urban underclass. Washington, DC: The Brookings Institution.

Wilson, R. (2011). Visualizing racial segregation differentlyexploring changing patterns from the effect of underlying geographic distributions. Cityscape: A journal of policy development and research, 13(2). U.S. Department of Housing and Urban Development, Office of Policy Development and Research.

Appendix

Table 1A. Missing U.S. Census Block Group Data by Tract Number, 1990, 2000 and 2010

Metro Area or Jurisdiction	1990	2000	2010
Richmond City	060798	N/A	N/A
Henrico County	201001, 200405, 200108	N/A	980100
Chesterfield County	100300, 100915, 100812	N/A	N/A
Jefferson County	012204, 011301, 011002, 009000, 008700, 011101, 011102	21111	21111
Mecklenburg County	000400	005606	980010, 980200
Hamilton County	00200	N/A	N/A

Figure 1A. School and residential percent change, Black-White/White-Black dissimilarity index, Chattanooga-Hamilton County, 1990-2010.

Source: NCES Common Core of Data, 1992-1993, 1999-2000, 2008-2009; U.S. Census 1990, 2000, 2010.

Figure 2A. Black students and residents exposure to White students and residents, four metropolitan areas, 1990-2010.

Source: NCES Common Core of Data, 1992-1993, 1999-2000, 2008-2009; U.S. Census 1990, 2000, 2010.

Cite This Article as: Teachers College Record Volume 115 Number 6, 2013, p. 1-45 https://www.tcrecord.org ID Number: 16988, Date Accessed: 3/16/2022 7:24:15 AM Purchase Reprint Rights for this article or review