The fracturing of Wake’s consensus had multiple causes, but it largely broke apart over two differing models of educational arrangements in American life. The first, a traditional model, presents schools as integral parts of communities (Reynolds 1999). Most children receive their education where they live, and the community’s schools, which often enjoy considerable autonomy from higher levels of government, become an important part of the fabric of social life as they prepare citizens for productive adult roles and democratic citizenship. The second and competing model views schools as having a broader purpose in society, one tied to democracy in a different way. Whereas in the early part of the twentieth century, schools were tasked with integrating millions of immigrants in the American way of life and preparing workers for various roles in the new industrial economy (Nasaw 1979), more recently they have become vehicles for promoting equal opportunity, an important foundation of meritocracy (Darling-Hammond 2010). Achieving the goals of this “diversity” model necessitates greater centralized control of education than is believed desirable by proponents of the traditional “neighborhood” model.

It is fair to say that much of the research has provided support to advocates of Wake’s established diversity policy. Some scholars have been very interested in the role race and income—or social class—play in explaining educational achievement. In the modern period, many have focused on the disruption of racial integration in schools by “white flight” and “bright flight,” where parents with means relocate children from city schools that often are failing to those, generally in the suburbs, that are more white, more middle class, and perceived to be more successful (Clotfelter 2001; Rossell, Armor, and Walberg 2002; Smrekar and Goldring 2009a). This has led to increasing residential segregation by both class and race, with attendant risks for the resegregation of schools and the reversal of the integration process with implications for all children, but especially for those who are minority and poorer (Boger and Orfield 2005; Clotfelter 2004; Orfield and Lee 2005; Orfield and Yun 1999). Orfield, Eaton, and the Harvard Project on School Desegregation (1996) argue that school desegregation has been dismantled through a variety of mechanisms, including courts’ unwillingness to create school boundaries that include sufficient numbers of middle-class children, policy makers’ reliance on funding rather than on integration strategies to improve education, and the failure to address the critical relationship between housing patterns and school segregation. Boger and Orfield (2005) maintain that despite numerous early court orders mandating integration, many schools in the South are now actually resegregating. The forces at work in this process include the removal of judicial pressure after systems are declared “unitary” or sufficiently desegregated; tremendous immigration, particularly from Latin America and Asia; and federal legislation such as No Child Left Behind, which places disproportionate administrative burdens on those schools that are performing poorly and already lack resources. Frankenberg and Orfield (2012) point to the suburbs as places where resegregation is occurring most dramatically.

Back in the 1960s, the influential Coleman Report (Coleman et al. 1966) found that disproportionately concentrating minority children in schools was associated with lower levels of achievement, outcomes primarily attributable to socioeconomic differences among the races. These findings strengthened researchers’ interest in determining whether it was the context of racial concentration that was consequential or whether differences in achievement were substantially owing to social class composition. Subsequent work has reinforced Coleman and his colleagues’ findings regarding the importance of income and social class (Darling-Hammond 2010; Grubb 2009; Kahlenberg 2007; Kozol 2005; Orfield and Lee 2005; Ryan 2010). Mickelson, Bottia, and Lambert (2013) demonstrate the importance of both racial and class heterogeneity in schools, and Gamoran (2001) argues that in the long term, school segregation by income will persist, even though racial integration may continue. Current research on the geography of school inequality uses national data to demonstrate that concentrations of poverty in schools are responsible for most of what otherwise might appear to be racial differentials in achievement (Logan, Minca, and Adar 2012). This research has profound implications for the Wake County case. A southern jurisdiction with a significant population of African Americans, Wake had a policy of diversifying schools along racial lines until 2000. As we shall see, when policies like this were deemed unconstitutional in other school districts, the county shifted gears to integrate schools using income and academic achievement—hence making them socioeconomically heterogeneous.

Just as schools with large poor or minority populations have been shown to struggle, social scientists from a number of disciplines have demonstrated that student bodies mixed by race, ethnicity, and family income perform quite well (Braddock and Eitle 2004; Hallinan 1988; Ladwig 2010; Linn and Welner, 2007; Mickelson, 2014; Rossell, Armor, and Walberg 2002; Vigdor and Ludwig 2008; Wells and Crain 1994; Welner 2006; see Lauen and Gaddis 2013 for cautionary evidence regarding this conclusion). Interestingly, while low-income youth from disadvantaged minority backgrounds benefit the most from diverse student bodies, middle-class white and Asian youths are also better off. For example, Rumberger and Palardy (2005) analyze the 1988 National Education Longitudinal Study data of achievement growth among high school students. They find that the socioeconomic characteristics of schools had important positive effects on the learning of both advantaged and disadvantaged students, independent of the individual student’s situation, with the institutional differences being a function of variation in teacher expectations, the amount of homework, course rigor, and students’ perceptions of campus safety. Students also benefit socially from attending integrated schools because the interpersonal relationships they build there provide a foundation for positive racial attitudes and patterns of social interaction as they grow up (Braddock and Gonzalez 2010; Pettigrew and Tropp 2004, 2006; Stearns 2010).

Social scientists have focused almost exclusively on the relationship between the composition of schools and student academic achievement. They have devoted much less energy to studying how public schools are organized in the first place. It is this that is the central focus of our work. There are some case studies of how politics and social change in large urban districts have impacted school organization. Ryan (2010), for example, traces the history of desegregation in Richmond schools and argues that boundaries segregating minorities in the city from middle-class whites in the larger county were consequential in the decline of urban schools in terms of achievement levels (see also Lleras 2008; Roscigno, Tomaskovic-Devey, and Crowley 2006). Cuban (2010) addresses similar matters in Austin, Texas, as do Mickelson (2001), Mickelson, Smith, and Hawn Nelson (2015), and Smith (2004) for Charlotte-Mecklenburg. Ravitch (2010) details the political and social forces that resulted in changes in school leadership and organization for New York City and San Diego, among others. Portz, Stein, and Jones (1999) show how differences in what they call “civic capacity” in Boston, Pittsburgh, and St. Louis affected the pattern and extent of changes in the way public schools were organized in those cities.

While these latter studies help to frame our investigation, they rely heavily on archival and interview sources for evidence. Our goal is to tie the Wake County case to theory in a number of social sciences and use a broader range of methods and data. We begin by discussing the concept of social capital and how it underpins our investigation of the diversity and neighborhood school models of student assignment.

The school is another level of social organization clearly relevant to our story. Children develop connections, a form of bonding social capital, to other students and teachers at their schools. Coleman (1990) argues these bonds facilitate learning and the acquisition of social norms.² As at home, connections among classmates and between children and teachers are more effective when they extend over time, promoting time closure. Scholars have already noted the difficulties schools experience when there is high teacher turnover, thus reducing another form of time closure (Darling-Hammond 2010; Haberman 2005; MacDonald 1999). In his study, Grant (2009, 120–21) argues that high teacher turnover in Syracuse, New York, represented a significant loss of school social capital; by contrast, the stability of Raleigh’s instructional staff represented a correspondingly positive asset. Time closure is also reduced when children frequently change schools, often because the family moves. The disruption threatens academic achievement and means that children must either reestablish valuable connections in another setting or, potentially, reduce the effort expended in creating new relationships because they anticipate yet more disruption (DeLuca and Dayton 2009; Grubb 2009).

Putnam’s (2000) “bridging” social capital is also important. It describes the connections that occur between settings, in this case the home and the school. For example, some scholars argue Catholic schools facilitate children’s achievement because a common religion for students and teachers creates bonds supportive of learning. Although some suggest the effects are modest when appropriate background controls are introduced (Alexander and Pallas 1985; Graetz 1990; Raudenbush and Bryk 1986; Willms 1985), several studies have shown material positive effects of attending Catholic schools (Coleman, Hoffer, and Kilgore 1982; Gibbins and Bickel 1991; Jensen 1986). There is some evidence that if social capital is a “stock of social goodwill created through shared social norms and a sense of common membership,” connecting a public school tightly to its immediately surrounding community is worthwhile (Furstenberg 2005, 810).

In addition, children benefit from the social connections that parents have with others such as neighbors, school personnel, and work colleagues (Crosnoe 2004; Dufur, Parcel, and McKune 2008; Johnson, Crosnoe, and Elder 2001; Parcel and Dufur, 2001a, 2001b). These connections can be useful in supporting socialization by creating a broader network that children can access as they integrate into the larger community. The stronger the connections tapping bridging capital, the greater the resources to which children have access. An important form of bridging social capital is “intergenerational closure,” which Coleman (1990) defines as occurring when parents know the parents of their children’s friends. This enables adults to pool resources in establishing and enforcing norms for children, something that is made easier in stable neighborhoods. The connections also speak to ties among individuals within communities, another form of bridging social capital. To be sure, scholars continue to debate whether social capital built at the community level has positive effects. Putnam (2000) and Murray (2012) worry that declines in social capital have an adverse impact on community functioning, while others find mixed results and question whether social capital at the community level has actually declined and, if it has, whether such changes are consequential (Portes 2000). Still, the notion that individuals are embedded in community networks as a function of frequent contact with and geographic proximity to them is an important one that may indeed be consequential for social outcomes, including academic achievement.

Finally, norms are an important form of social capital. They bind members of social groups together and reflect the values and goals that members hold (Coleman 1990; Portes 1998). We have already noted that much social science literature argues that having adequate numbers of affluent and/or white children in schools and classrooms can help distribute middle-class norms regarding achievement and college attendance to children of lower socioeconomic status who might not otherwise have sufficient exposure to them (Kahlenberg 2001). At the same time, middle-class children learn to deal with peers from a variety of class backgrounds, important preparation for their adult lives (Blau 2003; Gurin, Dey, and Hurtado 2002; Moody 2001; Wells, Duran, and White 2008). It is also often the case that poorer and minority children have such little reserves of social capital that placement in a school with peers from wealthier families, even if it is located far from where they live, may actually help them establish useful relationships.

It does seem as though the breakdown in the consensus over Wake schools was at least partially a product of diminished trust in others and the school system and its governing board. Some individuals, particularly whites, are opposed to integration policies and prevent their implementation in a systematic and thorough manner. The opposition stems from social distrust, which is frequently manifested in the form of “threat.” This is the idea that as minority racial groups grow and acquire greater resources and political power, spatially proximate whites, who generally make up or are of the same race as the community’s elite, respond by advocating policies that protect the status quo (Key 1949). As the size of this threat grows along with an expanding or increasingly powerful minority population, whites feel a greater need to protect existing arrangements (Blalock 1967; Blumer 1958; Bobo and Hutchings 1996; Giles 1977). They use tools of social control like public policy to do so (D’Alessio, Stolzenberg, and Eitle 2002). Economists posit that this is because the dominant group believes it will be able to consume less of a public good if others gain access to it (Alesina, Baqir, and Easterly 1999; Kruse 2005). An alternative but related argument is that members of majority groups feel entitled, and when minorities compete with them for social resources, they derive negative views about those of the other race (Kent and Jacobs 2004, 2005; Lopez and Pantoja 2004).

This feeling can be mitigated by what is called “contact.” Individual and direct interaction with members of the minority has been found to be important and can presumably assuage or counteract the attitudes of a majority-race individual evoked by threat (Allport 1954; Jackman and Crane 1986). The more frequent this contact, the lower the animosity toward members of the minority race, either because the direct interaction dilutes the racial fears of majority race members and evokes trust between individuals of different races or because those who are naturally more tolerant of individuals of other races actively seek out relationships with them (Marshall and Stolle 2004; Pettigrew and Tropp 2006; Sigelman and Welch 1993; Stolle, Soroka, and Johnston 2008). As a result, majority-race members who live in communities with large numbers of minorities might have fears reduced if they have significant contact with individuals of the other race (Dixon et al. 2010; Tropp and Pettigrew 2005; Welch et al. 2001). In this view, more intergroup contact will elevate trust, thus building social capital and civil society more generally (Rahn and Rudolph 2005; Rudolph and Popp 2010). As we show, issues of both threat and proximity appear to be relevant to the events in Wake.

Political trust is also important. Events in Wake County developed as the public’s trust in governmental and political institutions across the United States was declining markedly. As we shall see, Wake’s school board was distrusted by a growing collection of parents with all types of political views. Since people who hold governmental institutions in low esteem generally disdain liberal policies or government intrusion in civic society (Hetherington 2005), purposive racial and socioeconomic integration of schools was viewed with a particularly jaundiced eye. This was at a time when, because of economic conditions accompanying the Great Recession, parents were also likely to be concerned about their children’s futures.

Heightened partisanship and ideological division undermine political trust further. The national parties have demonstrated increased levels of interparty polarization and intraparty homogeneity over the past few decades (Abramowitz 2010; Levendusky 2009; Theriault 2008). ...