Non-Criminal Policy Approaches to Reduce Racial Inequalities in Crime and Justice
Eradicating racial disparities in the criminal justice system, as has been documented in part one of this report, necessitates eliminating racial differences in criminal involvement. To do that, it is necessary to eliminate the systemic racism that underlies racialized disadvantage and racial residential segregation. With that in mind, in this chapter we address the noncriminal policy approaches (i.e., in systems outside of the criminal justice system) that hold promise for reducing racial inequality in the justice system. In addition to addressing the underlying causes of concentrated disadvantage and segregation, we seek to identify promising sites of intervention in adjacent social policy systems or institutions that can perpetuate, compound, or mitigate racial inequality in and outside the criminal justice system, such as education, child welfare, and health systems.
The non-criminal justice topics covered in this chapter reflect complex public policy issues that have been explored extensively in the social science literature. Here the committee synthesizes some of this evidence using high-quality studies with rigorous methodologies, systematic reviews, and authoritative sources to highlight potential strategies that can mitigate the relationship between the drivers of inequality and racial inequalities in crime and justice, consistent with the Statement of Task. When the evidence is mixed or findings are based on studies with a correlational research design, it is explicitly noted. The chapter begins with a discussion of strategies to improve the material well-being of communities, such as through economic supports and environmental interventions, reviews the evidence on related public health approaches to reduce crime and victimization, and
ends with a discussion of individual system interventions, such as in child welfare, education, and health care.
IMPROVING THE MATERIAL WELL-BEING OF COMMUNITIES
The eradication of hyper-disadvantage is a long-term goal, one that policy makers need to immediately begin taking steps toward. In the meantime, certain steps might be taken to attenuate the negative consequences of the racialized patterning of social life in the United States through policies and programs to improve the material well-being of communities. This section addresses some possibilities toward that end, with a focus on economic and environmental interventions at the local level.
Policies and Programs to Address Underinvestment
This report takes a historical view on racial inequalities in crime and justice. It is a perspective that informs the need for structural policies and programs to address decades-long patterns of disinvestment (including underinvestment) from communities that have experienced disproportionately high rates of crime and criminal justice contact. The impacts of these policies and practices have been well documented by the social science literature (e.g., Faber, 2020; Lipsitz, 2011; Massey, 2007; Fullilove, 2001). For example, Faber (2020) traces the long-term consequences that redlining has had on segregation and patterns of spatial inequality, such as the enduring Black/White wealth gap. Benns and colleagues (2020) document an association between redlining and increased levels of gun violence in Louisville. The development of subsidized housing projects that concentrated low-income households in largely racially isolated groups has also been studied in relation to neighborhood effects and inequality (Freeman and Botein, 2002). For example, Popkin and colleagues (2000) use qualitative data to paint a portrait of public housing projects in Chicago (IL) in the 1990s, which were characterized by violence, crime, and hopelessness among residents as well as dysfunction in the Chicago Housing Authority. Massey and Kanaiaupuni (1993, p. 120) assert that public housing
…represents a key institutional mechanism for concentrating large numbers of poor people within a small geographic space, often within dense, highrise buildings. Because low-income projects were systematically targeted to Black neighborhoods in a discriminatory fashion…this institutional mechanism greatly exacerbated the degree of poverty concentration for one group in particular—blacks.
In short, historical patterns of housing policy, such as segregation and redlining practices, have had cascading consequences for a variety of contemporary outcomes (e.g., wealth, neighborhood demographics, segregation, and crime). Investing in communities with concentrated disadvantage and high rates of crime and victimization is one promising mechanism for addressing the harms perpetuated by historical policies and practices.
Broadly speaking, investments from private sources such as banks directed at fortifying a community’s housing stock can stimulate community social organization and help end the spiral of underinvestment and decline. As Vélez and Lyons (2014, p. 225) write, “External capital investments, ranging from bank lending to resources for public services, make or break the ability of neighborhoods to control crime.” This form of investment offers important potential for remedying the relatively higher levels of neighborhood crime in Black and Latino communities than in White ones.
Investments that contribute to homeownership have also been explored in relation to reducing crime and racial inequalities, as large gaps in neighborhood crime rates exist between White and minority communities (see Chapter 3). A small but important body of work points to the important relationship between home mortgage lending and crime, which offers some suggestive evidence about the race-crime link at the neighborhood level. Controlling for a number of neighborhood-level characteristics (e.g., socioeconomic disadvantage, residential instability, immigrant concentration, and prevalence of young males), Saporu and colleagues (2011) use a multilevel Poisson model to establish the relationship between residential lending and violent and property crime among neighborhoods of different racial composition. The authors find that “residential lending has the greatest payoff in areas that suffer the most from criminal conduct: African American, Latino, and more disadvantaged neighborhoods” (Saporu et al., 2011, p. 96). A study of Cleveland neighborhoods employs time-series Poisson models and finds that relatively large shares of home mortgage loan dollars were associated with a reduction in rates of intimate-partner violence, and this correlation is enhanced as neighborhood disadvantage increases (Boggess and Chamberlain, 2021).
Relatedly, a body of work largely concerned with the fallout of the Great Recession, including the housing crisis, has focused on the consequences of neighborhoods losing homeownership through foreclosures and evictions. When homeownership loss spreads across a community, the community becomes vulnerable to crime and related processes. Minority communities, given their disproportionate share of these housing losses,
become particularly vulnerable to public safety concerns. For instance, Kirk (2021) finds that neighborhoods with relatively large shares of evictions observed an increase in violent crime rates, and that such eviction levels help explain the increased risk of violent crime in Boston neighborhoods with relatively large shares of Black residents and high levels of concentrated disadvantage. Boessen and Chamberlain (2017) find that among Cleveland neighborhoods, being surrounded by foreclosures is associated with an increase in violent crime rates.
Some work has begun to explore what blocked access to lending capital means for neighborhood safety. In a study of Boston neighborhoods, Kirk assesses the geographic distribution of denials for home mortgage lending and the relationship between this patterning and violent crime. She finds that disadvantaged neighborhoods with large shares of Black residents have high rates of mortgage denials. Multivariate analyses also indicate that neighborhoods characterized by high levels of mortgage denials have higher levels of violent crime, partly because these denials increase resident perceptions of neighborhood social problems. Kirk (2021, p. 119) posits “that mortgage denials are a continuation of past acts of housing discrimination.”
One limitation posed by these studies is the correlational nature of the analysis and findings. There may be unobserved or unaccounted-for neighborhood characteristics that could confound the relationship between investments and crime rates, which makes it difficult for researchers to establish a causal relationship. One significant caution to apply in interpreting these research findings is that evidence suggests the housing crisis that drove the Great Recession was largely the result of programs that incentivized subprime loans and an increase in household debt (Mian and Sufi, 2015). Nevertheless, policies that encourage private investments by banks into communities, especially policies that provide resources for home purchase or home retention, require further study as they could provide an opportunity to curb racial inequality in neighborhood crime, especially violence. Negative externalities associated with losses of homeownership, like eviction and foreclosures, would also potentially be thwarted with such place-based solutions.
Public Capital Investments
Cities can provide matching grants to community-based organizations for neighborhood improvement projects or services, which have been linked to reductions in crime and by extension in racial inequalities because of the focus on disadvantaged neighborhoods (Ramey and Shrider, 2014). Ramey and Shrider (2014) find that matching funds from a Seattle city program were associated with reductions in violent crime for all neighborhoods, and that this form of public capital is strongly correlated with reductions
of violence in disadvantaged neighborhoods.1 While the Seattle program relied on community-based organizations to plan, organize, and execute the projects, it also built a coalition among public agencies and local organizations to help build capacity to address social control and crime prevention. In more disadvantaged neighborhoods with relatively high levels of cumulative funding, the authors estimate that the expected violent crime rate dropped from almost 40 violent crimes per 1,000 to just over 15 violent crimes per 1,000 in 2007. This contingent finding suggests that neighborhoods with relatively high levels of concentrated disadvantage benefit the most from this form of investment.
A second paper, by Shrider and Ramey (2018), continues to document the importance of public capital in relation to violence in Seattle neighborhoods. In particular it finds that neighborhood matching funds could help to reduce violence by increasing local levels of mortgage lending. This indirect relationship is pronounced in Seattle’s most disadvantaged neighborhoods. Neighborhood matching fund programs are an example of a type of community social control identified in Carr’s (2005) study of a neighborhood in Chicago. Carr found that people are often too busy or fearful to engage in direct interventions in local problems, so residents worked to address local problems by engaging in a new parochial social control based on partnerships between the people living in the neighborhood and public agencies. See Chapter 6 for a discussion of social control and collective efficacy as principles of community-driven safety and responses to harm.
Such public investments may foster long-term commitments to, and investments in, communities as residents see the successes of their tangible efforts, and this in turn might lead to subsequent reductions in crime and better neighborhood conditions overall (Krivo, 2014). Vélez and Lyons (2014) note that the documented relationship between the Neighborhood Matching Fund program and violent crime reduction in Seattle underscores the significant role that external investments can play in facilitating neighborhood well-being, social control, and crime. However, researchers caution that before forming partnerships, local organizations need to consider who they are investing in, who is being neglected, whether the investments are beneficial or harmful (e.g., subprime loans), and how the political
1 The Seattle Neighborhood Matching Fund was established to provide matching funds for neighborhood improvement, organization, or projects developed and implemented by community members. A central component is the community match, which requires recipients to match their award with contributions from the community (e.g., as volunteer time, materials donated, professional services donated, or cash). The program also contains a stipulation that at least 51 percent of the total Neighborhood Matching Fund funding must go to moderate-to-low–income neighborhoods and project sites. For more information see http://www.seattle.gov/neighborhoods/programs-and-services/neighborhood-matching-fund
context might affect everyone involved (Krivo, 2014; Leverentz, 2014; Vélez and Lyons, 2014).
Public-Private Partnerships to Promote Community Development
Cities and municipalities can collaborate with local businesses to fund Business Improvement Districts (BIDs). BIDs are self-taxing districts created by property owners to provide local public goods, typically for community service. BIDs allocate resources to enhance a variety of services, such as sanitation and public safety, which are typically provided by public agencies. In the case of BIDs, these services are funded by increased tax burdens on commercial properties that have been approved by local residents (Meltzer, 2012). A primary goal of BIDs is to improve the built environment as well as public safety; hence, they are often termed “clean and safe” initiatives (Hoyt, 2004). Studies document that this type of public capital intervention is linked to reductions in serious street crime (MacDonald et al., 2010; Brooks, 2008; Hoyt, 2004).
MacDonald and colleagues (2010) find that BID implementation was associated with a 12 percent reduction in robberies and an eight percent reduction in total violent crimes in Los Angeles neighborhoods. Brooks (2008) finds that neighborhoods with BIDs in Los Angeles have between six percent and 10 percent less crime than otherwise similar neighborhoods. Additional research indicates that the introduction of BIDs in Los Angeles was associated with a decrease in crime and police arrests over time, with a corresponding increase in criminal justice cost savings and minimal displacement of local crime to neighboring areas (Cook and MacDonald, 2011). However, Clutter and colleagues (2019) find that in Cincinnati, BIDs played a role in the spatial distribution of robberies at the level of street blocks. That is, street blocks within the BIDs observed higher robbery counts, after controlling for street factors. More research is needed in different neighborhood contexts to parse the mechanisms or mediating factors that might influence the relationship between BIDs and violent and nonviolent crime.
Shaping the Built Environment
As noted in Chapter 3, crime can be influenced by the built environment as well as the socioeconomic environment, and both differ across racial groups. Broadly, the built environment reflects the physical conditions of a community, including human-made physical components, design, and permitted use of space, including transportation, parks, and open space
(the National Academies of Sciences, Engineering, and Medicine, 2017). Place-based interventions that modify features of the built environment have been the subject of experimental studies to examine how crime might be reduced in communities.
One such intervention is the “greening” of spaces, which entails the remediation of vacant lots or spaces by removing trash and debris, grading the land, planting trees and grass, and/or installing fencing. A review by Kondo and colleagues (2018) indicates that housing and blight remediation of buildings and land are the interventions that most consistently reduce violence, as compared to interventions addressing alcohol availability, transit changes, and school opening. Another systematic review, which covers 10 empirical studies, finds that while there is evidence to support the positive impact of green space on crime, comparisons across the studies are difficult due to variations in study design and mixed measurement and results (Bogar and Beyer, 2016). Shepley and colleagues (2019) synthesize some of the quantitative and qualitative evidence on the impact of green space on violent crime (i.e., on murder and non-negligent manslaughter, forcible rape, robbery, and aggravated assault). Their review of 45 publications indicates that a limited evidence base suggests that the presence of parks and green space reduces urban crime; however, many papers paid insufficient attention to confounding variables. Among the papers included in their review, Boessen and Hipp (2018) examine nearby demographics as moderators and find some protective influence of parks on Latino neighborhoods. Shepley and colleagues (2019) also point to a few hypothesized mechanisms by which green space might influence crime: social interaction and recreation, community perception, nature-induced stress reduction, climate modulation, and spaces with territorial definition (Shepley et al., 2019).
Recent research identifies the demolition of vacant or abandoned buildings as a promising strategy to reduce crime in the immediate vicinity, but effects are limited spatially (Wheeler et al., 2018). Other systematic reviews reinforce these findings on the association between modifying the built environment and reductions in crime (e.g., Inlow, 2021). MacDonald and Stokes (2020) trace reductions in crime to gentrification and subsequent changes to land use but note the need for a clearer analysis of how crime becomes spatially displaced in these processes. While these studies do not examine racial disparities in criminal victimization or offending as a primary outcome, there is a body of literature that suggests there are racial and socioeconomic inequalities in access to green spaces across the United States (Nesbitt et al., 2019; Wen et al., 2013).
Communities, particularly those with vacant lots, abandoned houses, and distressed commercial corridors, could consider strategic investment in these types of place-based programs. There is evidence to suggest that such programs are scalable, are cost effective, and reduce crime (MacDonald et
al., 2021; Kondo et al., 2015). They could be concentrated in neighborhoods with the highest rates of crime and victimization, could be replicated across the country, and could employ individuals from the communities where they are applied. In seeking to improve public safety, communities need to examine not only how to “restore people, but also restore places” (MacDonald et al., 2021).
It’s also important to note that revitalization initiatives can have unintended consequences, including gentrification. Undertaking such work is a balancing act. On the one hand, concentrated poverty can have a pernicious effect on communities, but on the other hand, gentrification can result in people being priced out of and displaced from their communities. Therefore, communities need to be mindful of these unintended consequences when undertaking such initiatives. See Box 7-1 for an example of a community urban greening program with promising results in crime reduction at relatively low cost.
Moving to Opportunity
The Moving to Opportunity (MTO) project exemplifies an important place-based study that randomly assigned families to rent-subsidized housing vouchers to explore the effects of concentrated poverty and neighborhood conditions on life outcomes. In 1994, more than 4,500 low-income families were randomly assigned to one of three groups: (1) those offered a housing voucher that could be used to move to a low-poverty neighborhood, (2) those offered a traditional Section 8 housing voucher, and (3) a control group. The MTO project featured a series of follow-up evaluations with the families—once for an evaluation four to seven years after a given family had been assigned (e.g., Ludwig, Duncan, and Hirschfield, 2001; Ludwig et al., 2001; Ludwig, Duncan, and Pinkston, 2000), and again from 2008 to 2010 for an evaluation 10–15 years post assignment (e.g., Ludwig et al., 2013, 2012; Sciandra et al., 2013). These evaluations examined the effects of the MTO project on adults and children, how these effects changed over time, and the potential mechanisms by which the outcomes were produced. The main categories of outcomes that were studied over time were adult economic self-sufficiency, mental health, physical health, education, and risky behavior (i.e., crime, delinquency), all of which have been documented in the evaluative literature cited above.
With respect to the crime and delinquency outcomes, the offer to relocate to lower-poverty areas was found to reduce arrests among female youth for violent and property crimes, relative to the control group. For males, the offer to relocate reduced arrests for violent crime, at least in the short run, but increased problem behaviors and property crime arrests. Ludwig, Duncan, and Hirschfield (2001) find a 30–50 percent reduction in juvenile arrests for violent offenses among the group of relocated families. Sciandra and colleagues (2013) document substantial initial reductions in violent-crime arrests along with sizable increases in property-crime arrests for experimental group males. These crime effects attenuate over time along with differences in neighborhood conditions.
The gender difference in treatment effects seems to reflect differences in how male and female youths from disadvantaged backgrounds adapt and respond to similar new neighborhood environments (Kling et al., 2005). Chetty and colleagues (2016) examine the long-term effects of MTO and find differential effects depending on the age of the children when they moved. Children who moved when they were younger (i.e., before the age of 13) saw increases in college attendance and earnings, while children who moved during adolescence experienced slightly negative effects. These findings corroborate the child development literature that emphasizes the importance of early intervention to shape life outcomes (the National Academies, 2019a).
The randomized nature of the MTO project has made a major contribution to the social sciences, as the authors were able to eliminate selection bias, a significant confounding variable in this area of research. However, despite the strong experimental design of the studies, scholars debate the generalizability and application of the findings to neighborhood-level theory broadly (Sampson, 2008). For example, the MTO project was an individual-level intervention for primarily Black families living in extreme poverty, which makes the findings limited to a specific segment of the population. Furthermore, the social mechanisms that explain the neighborhood effects observed in the MTO project are unknown, given the many correlates to neighborhood poverty (Sampson, 2012). Nonetheless, the MTO project provides important findings for reducing racial inequalities and directions for future research that integrates other important factors such as urban dynamics, social structure, and developmental neighborhood effects.
Labor Market Solutions
There is a well-developed literature on the relationship among race, labor markets, and crime (Lang and Kahn-Lang Spitzer, 2020; Crutchfield, 2014; Bushway and Reuter, 2002; Western et al., 2001). Wilson’s (1996) theory of joblessness postulates that poor job prospects drive crime rates through the weakening of social ties, producing psychological strain, and through a reduced perceived cost of engaging in crime. This theory has been reinforced by research indicating that joblessness has a stronger effect on neighborhood violent crime than poverty itself (Dollar et al., 2019). Labor market discrimination has also been the subject of research, based on job application data, as differential treatment by race appears to still be prevalent (Bertrand and Mullainathan, 2004), particularly in areas with high crime rates (Mobasseri, 2019). In addition, qualitative research corroborates these findings of practices of racial discrimination in the labor market, particularly toward job applicants who are Black and/or from low-income backgrounds (Kirshenman and Neckerman, 2019). Labor market solutions offer one potential lever within a broader context of cumulative, intergenerational economic disadvantage for specific racial and ethnic groups (e.g., Black, Latino, and American Indian populations). This economic disadvantage can be traced in part to the patterns of policy investment in White populations that many other racial and ethnic groups did not benefit from (see Chapters 1 and 3 for discussions on structural racism and concentrated disadvantage, respectively).
This section explores labor market solutions as levers to provide opportunities for people living in disadvantage or those returning from incarceration to secure gainful employment. It is important to make a clear distinction between recommended efforts to improve the labor market prospects of
adults living in disadvantaged neighborhoods and efforts to enhance the reentry of previously incarcerated persons. Both efforts are thought to be anti-criminogenic. However, while the former aims to improve the material well-being of disadvantaged communities and their residents and thereby reduce crime rates and racial inequalities, the latter aims to improve job prospects during reentry from prison, more specifically to reduce recidivism.
Employment Opportunities for Labor Market Success
Chapter 4 documents the consequences of criminal justice system involvement for racial inequality. One significant consequence is the impact that system involvement can have on future employment opportunities. While incarceration is often studied as a barrier to securing employment, research shows that even low-level offenses and/or history of arrest or conviction without incarceration can affect employment outcomes negatively (Agan and Starr, 2018; Uggen et al., 2014).
The effect of having a record of incarceration as a major barrier to securing employment is pronounced among Black job seekers (Decker et al., 2015). Moreover, Black people who have returned from prison have higher unemployment and recidivism rates when compared to White people returning from prison (Lockwood et al., 2015). The neighborhood conditions that a person who is formerly incarcerated returns to can also shape labor market outcomes (Schnepel, 2018; Yang, 2017; Morenoff and Harding, 2011; Raphael and Weiman, 2007). For example, returning to communities with high unemployment rates is a risk factor for violent recidivism among Black men (Wang et al., 2010). Given that unemployment is a strong predictor of recidivism, programs or policies that enhance employment opportunities in support of reentry efforts are a potential mechanism for reducing racial disparities in criminal just involvement (the National Academies, 2022a).
Programs and conditions designed to promote labor market success have been studied as interventions to support reentry efforts. Previous research indicates that programs aimed at shifting people in low-income settings from criminal activity and into employment need to reduce the attraction of crime and treat substance abuse problems, while providing social and educational supports to help high-risk individuals obtain gainful employment (Bushway and Reuter, 2002). Quality social ties (e.g., familial connections) have been shown to influence recidivism and job attainment (Berg and Huebner, 2011). Zhang and colleagues (2019) conducted a community-level review of reentry programs and found that programs that provide training and placement services to returning citizens using a holistic approach (i.e., with a focus on training and job placement) are the most effective in ensuring successful reentry into the workforce.
Research emphasizes the importance of placement into high-quality jobs with potential for upward mobility through educational (vocational and GED [General Educational Development Test]-based) and entrepreneurship programs (Zhang et al., 2019; Crutchfield et al., 2011). Lacoe and Betesh (2019) find that the long-term effectiveness of employment-focused approaches to improve employment outcomes or to reduce criminal justice system involvement is not consistent across studies; nonetheless, their review reinforces the value of vocational training as a prerelease intervention. Two randomized controlled trials (RCTs) suggest that transitional employment programs are not as effective beyond the initial implementation period and therefore have less success in terms of long-term employment outcomes (Cook et al., 2015; Valentine and Redcross, 2015).
In summary, a better understanding of the key mechanisms of reentry employment programs and contextual, protective factors to support sustained labor market success would inform a strong evidence base from which to build solutions. Box 7-2 presents an example of a policy intended to remove barriers to employment, for which evidence suggests there might be unintended consequences.
Homeboy Industries: Community-Building Reentry Services
Homeboy Industries is one example of a promising program that provides job training opportunities and free social services for people who are formerly gang involved and/or previously incarcerated. Each client is assigned a case manager who helps him or her set a goal plan, which might be obtaining a high school diploma or GED, following a course of action for release from parole or probation, or identifying needed services, classes, and skills in order to find outside employment. Many of the participants arrive into young adulthood with no support system and absent families. Homeboy Industries provides a supportive community, a sense of family, and a place where they can come to help find their strengths, learn job skills, get an education, learn new life skills, and become contributing members to their families and communities (Office of Justice Programs, 2021a). The program’s efforts to transform the lives of its participants reflect an appreciation of the social nature of personal transformation and redemption, with a focus on building the capacity of its participants. In contrast to the “focused deterrence” framework, Homeboy Industries’ efforts, described by sociologist Edward Flores (2013) as “integrative redemption,” are shaped by a theory that emphasizes “recovery by way of reintegration into the mundane world.”
Despite the challenges of evaluating the success of such a program within a dynamic environment, Leap and colleagues (2010) employed quantitative and qualitative methods to assess the program and identified the
following mediators that shape program outcomes: improved self-esteem, attachment, self-efficacy, emotion regulation, coping skills, and prosocial orientation. In 2020, the program documented having 476 full-time trainee participants, 2,746 additional community members receiving program and service support, and $41,400 in fines and fees removed for participants
(Homeboy Industries, 2020). Taken together, community-building and reentry programs such as Homeboy Industries and Cure Violence (discussed in Chapter 6) share a set of practices that encourage change. Jones (2018) notes that these programs appreciate “the importance of place, time, and community as well as the power of transformational relationships.”
Crime Control Approaches through Employment Programs
The literature on jobs as a delinquency prevention strategy for youth is mixed in terms of practical application (Millenky et al., 2011; Schochet et al., 2008; Uggen, 2000; Bloom et al., 1997). It is worth noting, however, that there have been promising results from summer youth employment programs (Heller, 2021). For example, experiments in Chicago, New York City (NY), and Boston (MA) generally have led to similar findings, namely that summer youth employment programs resulted in large declines in criminal justice involvement and violence, although little improvement in future employment was found on average (Kessler et al., 2021; Davis and Heller, 2020; Modestino, 2019; Gelber et al., 2016; Heller, 2014). However, using a machine learning method Davis and Heller (2020) uncover some heterogeneity of effect: youth who benefit most from the summer employment program in Chicago are more likely to be younger, Hispanic, and female; more engaged in school; and less criminally involved than their counterparts, who are not predicted to experience positive outcomes. These data on subgroups could inform tailored approaches to employment programs for youth.
The Role of Public Transportation Systems and Spatial Mismatch in Labor Markets
A labor market problem for working class and low-income people in some locations is the spatial mismatch between where the jobs are and where potential employees live. This is part of the explanation for Black/White employment differences (Gobillon et al., 2007; McLafferty and Preston, 1996; Holzer, 1991). Historically, Black Americans and Latino people have tended to live in central cities, frequently in highly segregated neighborhoods (see Chapter 3), but increasingly industries and jobs have dispersed out of the central city into suburbs, thus contributing to racial labor market inequality. McLafferty and Preston (1996) find evidence that supports this thesis and note that African American and Hispanic women’s experiences with mismatch result from their being more reliant on public transportation. However, it should be noted that more recent trends show that the suburbs are diversifying and that Black and Latino households in diverse suburbs have higher median incomes than the national average (Rastogi, 2021). Hellerstein and colleagues (2008) believe that spatial
mismatch is less the problem than what they refer to as “racial mismatch,” the proximity of jobs where people of color are hired to where people of color live. However the mismatch problem is conceptualized, it remains the case that residential segregation, combined with job and industrial dispersion, perpetuates Black/White and Hispanic/White employment inequalities.
Improving the routing and efficiency of public transportation may be another area of research to address this spatial mismatch between workers from disadvantaged neighborhoods and the locations of jobs. Workers need access to affordable, efficient transportation, as public transit that requires residents of poor neighborhoods to make multiple transfers and travel for hours between home and work is unlikely to improve their labor market prospects.
If designed and maintained properly, public transportation facilitates safe mobility and is accessible to all residents, regardless of geographic location. Moreover, transportation systems can play a role in mediating spatial disadvantage by providing residents with access to economic opportunity, social goods, and community services (Dodson et al., 2004). However, current research suggests that transportation costs are a barrier to mobility for households in poverty, which are disproportionately represented by African American and Hispanic people (Federal Highway Administration, 2014).
In addition, Andersson and colleagues (2018), assessing how improved job accessibility where a spatial mismatch is present might affect unemployment, find that it significantly decreases the length of unemployment among lower-paid, displaced workers. Their findings also revealed that Black workers, women workers, and older workers were more sensitive to the effects of job accessibility than the other subgroups studied. Although the evidence presented here does not directly connect transportation and job accessibility to racially disparate outcomes in criminal behavior and/or system involvement, this section presents a potential area for future research: if and how racial inequalities in crime and justice might be addressed given the disparities that are observed in job accessibility and access to quality transportation.
PUBLIC HEALTH APPROACHES TO VIOLENCE
Public safety and violence are significant indicators of health and community well-being. Violent victimization affects health by causing psychological and physical injury, which can lead to disability and, in some cases, premature death. Beyond the risk of injury and death, violent victimization also has far-reaching health consequences for individuals, families, and neighborhoods. Furthermore, research shows that simply being exposed to violence can have detrimental effects on physical and psychological wellbeing (Pinderhughes et al., 2015; Felitti et al., 1998). Violent victimization
and exposure to violence have been linked to poor health outcomes among racial minority groups, including chronic diseases (e.g., ischemic heart disease, cancer, stroke, chronic obstructive lung disease, diabetes, and hepatitis), asthma-related symptoms, obesity, posttraumatic stress disorder, depression, and substance abuse (Prevention Institute, 2011). Some research also indicates a link between neighborhood crime rates and adverse birth outcomes, such as preterm birth and low birthweight (Egerter et al., 2011).
Given the connection between public health and violence, researchers have explored interventions that target unsafe environments to advance health and safety while also minimizing the harms of contact with the criminal justice system. Public health frameworks for addressing violence include efforts to advance data-driven policies that create environments that are less conducive to violence or that facilitate social conditions that constrain violence (Webster, 2022). The sections below discuss promising areas of public health policy, including firearm reduction policies, environmental exposure programs, and alcohol policies as they relate to racial disparities in violence, crime, and system contact. These are complex public policy issues that require careful consideration of the evidence, including study methodology, limitations of the available data, tradeoffs and unintended consequences, cost, and political will. The committee is not constituted to delve deeply into these issues or to provide recommendations in these areas; however, these areas warrant discussion as they are connected to racial inequalities in crime and justice. Here, the committee relies on systematic reviews, high-quality studies, and reviews of the literature from authoritative, evidence-based sources to synthesize some of the literature on public health approaches to violence.
Firearm Violence and Related Policies
To address the problem of racial disparities in crime, victimization, and criminal justice involvement, one needs to consider the issue of gun violence. As Chapter 2 discussed, the United States has observed a significant increase in firearm homicides in recent years. Gun violence is a racial equity issue, as marginalized racial and ethnic groups are disproportionately harmed by gun-related injury and death, by arrests and criminal charges, and by diminished community cohesion as a result of neighborhood-level gun violence (see Chapters 2 and 3 for more on firearm violence trends). Johns Hopkins Center for Gun Violence Solutions has developed a racial equity impact assessment tool to identify and assess the factors influencing racial equity that should be considered with firearm policy development and implementation (Educational Fund to Stop Gun Violence et al., 2022). The Center proposes a racial equity framework for gun violence prevention and recommends the following for a more equitable gun violence prevention movement:
- Identify and engage diverse members, and routinely ask who is missing and invite them to join;
- Identify the collective interest and allow for all collective members to provide ongoing input in developing a shared vision and goals;
- Commit to personal and collective growth and healing (i.e., truth and reconciliation);
- Prioritize inclusivity when identifying priority policies, programs, and strategies that address interests of all communities represented;
- Identify stakeholders, their resources, and influence to more effectively reach key policy decision makers;
- Engage in ongoing evaluation that is inclusive of all members;
- Make equitable collaboration sustainable; and
- Celebrate the victories and acknowledge the inevitable setbacks.
Previous reviews have examined firearm violence and policies (see, for example, American Psychological Association, 2013; NRC, 2005; Kaplan, 1984); this section highlights two more recent, comprehensive reviews of the evidence.
The John Jay College Research Advisory Group on Preventing and Reducing Community Violence (hereafter the John Jay research group) underscores the need to confront the issues of firearms as a comprehensive approach to reducing violence. Firearm availability is associated with rates of homicide across the United States (Siegel et al., 2014; Cook and Ludwig, 2006) and is also a reliable predictor of rates of fatal police violence (Hemenway et al., 2019). The John Jay research group reviews the evidence on reducing access to firearms as one potential mechanism to reduce homicides and other violence. For example, one recent study examined changes in state-level policies governing access to firearms, including child-access prevention, right-to-carry laws, and stand-your-ground laws (Schell et al., 2020). The results suggest that child access protection laws alone could lead to 11 percent fewer firearm deaths. Other research points to purchaser licensing laws with comprehensive background check requirements as being associated with lower firearm homicides and suicide rates in one state (McCourt et al., 2020).
In summary, the John Jay research group concludes that violence can be reduced through policies that limit access to firearms, increase restrictions for people with histories of violent crime, reduce access to firearms for young people, impose waiting periods, and increase required firearm training (John Jay College Research Advisory Group on Preventing and Reducing Community Violence, 2020). A 2020 review of the evidence, conducted by a collaborative of researchers as part of the RAND Corporation’s Gun Policy in America initiative, examines the effects of firearm policies and how to improve the evidence on how such policies influence outcomes. The
authors assessed the empirical literature on 18 categories of firearm policies related to eight different outcomes, including suicide, violent crime, unintentional injuries and deaths, mass shootings, officer-involved shootings, defensive gun use, hunting and recreation, and the gun industry. While the review does not offer findings specifically on race and ethnicity, the authors do find evidence to support the conclusion that child-access prevention laws (i.e., safe storage laws) reduce self-inflicted firearm injuries and deaths. They also report moderate evidence to suggest that laws prohibiting gun ownership for individuals subject to domestic violence restraining orders can decrease intimate partner homicides, and waiting periods can reduce firearm suicides as well as total homicides.
More research is needed to provide evidence-based policy guidance that takes into account political feasibility, racial equity considerations, and implications for Second Amendment rights. The RAND review supports this notion with a recommendation to Congress to consider appropriating significant funds for research on firearm policy and violence reduction at levels commensurate with other public safety and health investments.
Reducing Harmful Environmental Exposures
Chapter 3 draws a connection between concentrated disadvantage and toxic exposures such as lead and air pollution, which have been linked to crime and criminal justice contact. This section describes some of the literature that connects lead exposure as one toxic mechanism for influencing criminal outcomes and as an environmental pathway that contributes to racial inequalities (Sampson and Winter, 2016).
Lead exposure early in life has been shown to increase learning deficits and impulse control problems, which can increase the risk of criminal offending over the life course (Muller et al., 2018). Early studies have documented an association between lead exposure and impulsivity, violent behavior, aggression, and delinquency (Needleman et al., 2002, 1996; Byers and Lord, 1943). While most of the early literature provides cross-sectional analyses, more recent research provides longitudinal data showing a positive correlation between lead exposure and different measures of delinquency or criminal outcomes (e.g., school suspensions, antisocial behaviors, arrests; Reyes, 2015; Amato et al., 2013; Wright et al., 2008; Dietrich et al., 2001). Lead risk has been found to be associated with incarceration in adulthood, a risk that was found to vary substantially at the census tract neighborhood level (Manduca and Sampson, 2021). Children who live closer to busy roads within a neighborhood are more likely to have high blood lead levels, which in turn have been linked to increased likelihood of school suspension and juvenile incarceration (Aizer and Currie, 2019; see the section below on Enhancing
Opportunities for Youth Learning and Development for more on school disciplinary practices).
Muller and colleagues (2018) provide a review of the social causes and consequences of lead exposure, including future aggression and criminal offending, and conclude that “addressing the problem of lead exposure entails turning our attention away from explanations based on character deficiencies and toward durable investments” to repair harm done by harmful environmental exposures (p. 275). For example, early life interventions, such as lead remediation, nutritional assessment, medical evaluation, and developmental surveillance, have been shown to reduce negative outcomes due to lead exposure (Billings and Schnepel, 2018). Collective efficacy and community mobilization can catalyze the deployment of environmental interventions to reduce toxic exposures, thereby reducing crime and criminal justice contact (see Chapter 6 for discussion about collective efficacy).
Alcohol Outlets and Community Violence
Researchers have examined the relationship between alcohol outlet density—the number of physical locations where alcoholic beverages are sold either per area or per population—and neighborhood-level violence as a possible site of intervention to make communities safer. Communities characterized by concentrated poverty are also more likely to have both higher alcohol outlet density and higher rates of violence (Furr-Holden et al., 2019; Trangenstein et al., 2018). Spatial patterns of violence in cities show a clear relationship with public or off-premise alcohol consumption, regardless of racial composition (Furr-Holden et al., 2019). However, the nature of racial inequality in concentrated disadvantage imposes a disparate impact on primarily Black and Latino neighborhoods. The documented relationship between alcohol outlet density and violent crime has prompted researchers to explore policy solutions to mitigate this relationship (e.g., zoning policies). One cost-effectiveness analysis found that reducing alcohol outlet density in one mid-Atlantic city by one quintile is associated with a decrease of 51 homicides per year and 764 disability-adjusted life years (Trangenstein et al., 2018).
Policies that restrict the sale of alcohol in bars or restaurants during specific hours of the week have been studied as an intervention to reduce violence. Biderman and colleagues (2010) detected a 10 percent homicide reduction associated with legal provisions governing alcohol sales. The phased repeal of laws in Virginia that once prohibited the sale of packaged liquor on Sundays provided an opportunity for Heaton (2012) to examine that repeal; it led to the finding that increased alcohol sales were correlated with increases in serious crimes—including violent crime—by 10 percent. Similarly, using county-level variations in Kansas laws governing the sale
and on-premises consumption of alcohol, Anderson and colleagues (2018) find that a 10 percent increase in the number of establishments licensed to sell alcohol by the drink increased violent crime by three to five percent. In a study of Chicago zoning codes, Twinam (2017) uses an instrumental variable strategy to find that in neighborhoods without high residential population density, the presence of liquor stores and late-hour bars was associated with higher levels of violent crime.
The Centers for Disease Control and Prevention’s Task Force on Community Preventive Services conducted a systematic review on the effectiveness of limiting alcohol outlet density to reduce excessive alcohol consumption and alcohol-related harms, including violence (Campbell et al., 2009). The authors concluded that cross-sectional studies reveal a positive association between alcohol outlet density and crime and violence. Based on this evidence, the Task Force recommends “the use of regulatory authority (e.g., through licensing and zoning)” to limit alcohol outlet density and related harms (Task Force on Community Preventive Services, 2009).
INTERVENTIONS IN OTHER SYSTEMS
This section explores interventions that take place in the adjacent social policy systems or institutions—such as education, child welfare, and health systems—that can perpetuate, compound, or mitigate racial inequality in and outside of the criminal justice system. The strategies discussed here take a life-course approach with the understanding that the cumulative nature of racial inequality is intergenerational for families and children. The overarching principles of these interventions are:
- to improve systems to strengthen families and protect children from system involvement;
- to enhance opportunities for youth learning and healthy development; and
- to promote health and well-being.
Another central tenet of these strategies is the recognition of the role that biases and institutional racism can play in shaping outcomes for youth and families.
Without reforms that address the underlying mechanisms of inequality and structural racism, there is a risk of shifting the responsibility for addressing social problems (e.g., violence or poverty) onto non-criminal justice systems while perpetuating racial inequality. In short, the mechanisms of inequality operate in systems outside of the criminal justice system as well. For example, the social psychology, health, and education literature shows that explicit and implicit bias can operate in education and health care
systems (Warikoo et al., 2016; Okonofua and Eberhardt, 2015; Paradies et al., 2014; IOM, 2003).
Strengthening Families and Protecting Children
Part I of this report documents how issues related to parenting and family stability, such as the loss of a child’s relationship to a parent, can make children more vulnerable to later criminal justice system involvement. It also documents how the criminal justice system has been a primary strategy for addressing problems faced by children, parents, and families, bringing more parents into contact with the criminal justice system. Non-White families can be disproportionately likely to face criminal justice system consequences for family-related issues. Both non-criminal justice system inequalities and inequalities in criminal justice system referrals make non-White parents and children more likely to face criminal consequences. And as Part I also shows, criminal legal system involvement tends to worsen outcomes for children and families. The strategies outlined in this section address needs related to families and children without relying on criminalization or criminal consequences.
Family Preservation/Reunification in Child Welfare
Children of color and those from low-income households are disproportionately referred to the child welfare system and, once referred, are more likely than other children to be placed out of the home either temporarily or permanently.
An ongoing challenge in the child welfare field has been managing the potential conflict between collaborating with families to prevent out-of-home placement or reunifying children with their families and, at the same time, addressing safety concerns. Federal laws such as the Adoption and Safe Families Act and amendments to the Child Abuse Prevention and Treatment Act called for a better balance between protecting children and preserving families. More recent legislation and policy guidance have sought to refocus program and practice efforts on ensuring child well-being and family connections as well as attending to safety and permanency. The passage of the Family First Prevention Services Act (FFPSA) in February 2018 acknowledges the positive outcomes families and communities experience when child welfare systems devote more resources to preventing the need for foster care. Previously, resources were targeted to families who have had their children placed in foster care, but very few resources are available to strengthen families and prevent foster care. The FFPSA provides funding to states to increase such family preservation services.
Family preservation services are short-term and family-focused, designed to assist families in crisis by improving parenting and family functioning
while keeping children safe. This approach grew out of the recognition that children need a safe and stable family and that separating children from their families is traumatic for them, often leaving lasting negative effects. Family preservation services vary by state but often include case management, individual and family therapy, and parenting skills training, as well as support services and referrals for families to community providers if they need help finding a job, securing housing, or accomplishing any other goal. Examples of evidence-informed intervention programs to keep children safe and promote permanency include the Nurse-Family Partnership, Safe Environment for Every Kid (SEEK), the Triple P Positive Parenting Program, and the Parent-Child Assistance Program (Casey Family Programs, 2018).
For those families whose preservation is not viable, kinship care may benefit children. Children who live in kinship families benefit from the ability to live with relatives when their parents cannot care for them. Research shows many benefits of kinship care. Compared to children in the general foster care population, for example, children in kinship care are better able to adjust to their new environments, less likely to experience behavioral and psychiatric problems, and less likely to change schools.
However, kinship families need more resources to support opportunities for the children in their care. Caregivers in kinship families may struggle, and many are grandparents living on fixed incomes. These caregivers may not have the financial resources to raise a child unexpectedly, especially children who face challenges because of their experience with abuse, neglect, and separation from their parents (Pittman, 2014). Programs and assistance to ensure that kinship caregivers access the benefits they are eligible for, such as Temporary Assistance for Needy Families, can help provide financial stability. Kinship caregivers tend to have higher poverty rates, less access to health insurance, and more physical or mental disabilities than other parents. Enhancing other community-based and government responses, such as promoting stable housing, ensuring that caregivers have health care and affordable legal representation, and easing barriers to school enrollment, could benefit both these children and their kinship families (Pittman, 2015).
Protecting Ties between Incarcerated Parents and Their Children and Families
Parental incarceration is one reason children may be fast-tracked to adoption under the Adoption and Safe Families Act, which requires states to file for the termination of parental rights for any child in foster care for 15 of the most recent 22 months (Children’s Bureau, 2021). According to the Children’s Bureau (2021),
…with the average prison sentence spanning more than 1 year—attributed in part to lengthy mandatory minimum sentences for common, nonviolent offenses—this requirement can be a significant barrier to reunification for incarcerated parents.
Some states’ child welfare statutes, such as those in Ohio and Iowa, specify the length of parental incarceration as grounds for termination of parental rights (Office of the Assistant Secretary for Planning and Evaluation, 2021). States like New York and Washington have taken steps to allow for increased flexibility regarding placement for children of incarcerated parents (Walsh, 2016).
Beyond issues related to legal status and parental rights, sustaining a relationship between an incarcerated parent and their child can be difficult due to a number of factors, not limited to visitation restrictions, difficulty navigating the criminal justice system, transportation challenges, and lack of consistent communication. Visitation rules for incarcerated individuals, though they vary across the United States, are an example. Arranging visits can be challenging as facilities may require pre-clearance, approved adult supervision for children under 18, and proof of identification for visitors (e.g., birth certificate or driver’s license; Annie E. Casey Foundation, 2011). Depending on where the parent is incarcerated, visitation may also require the child to be transported a significant distance to reach the facility, thus requiring reliable transportation from a supervising adult. Moreover, incarcerated individuals face significant communication barriers as they are not able to accept incoming calls and they must pay for all outgoing calls, often at significant cost. Without easy, consistent access to a phone and affordable calls, it can be difficult for an incarcerated parent to maintain consistent communication with their child. Providing incarcerated parents with a family conferencing model “can be extremely helpful in ensuring engagement and smoother communication between the parent, child, and child welfare agency” (Annie E. Casey Foundation, 2011).
Providing incarcerated parents with opportunities to participate in parenting courses can promote strong ties with their children and families. Parenting classes can “provide parents the opportunity to expand their understanding of child development and learn and practice effective and appropriate communication, discipline, problem-solving, and other parenting techniques” (Peterson, Cramer, and Fontaine, 2019; Peterson et al., 2019). Parenting Inside Out2 is one such program that uses parenting coaches to teach both mothers and fathers involved in the criminal justice system. In 1998, the New Hampshire Department of Corrections established the
Family Connections Center,3 which provides a variety of services and programs to incarcerated parents, including parenting education classes, parenting support groups, and healthy relationship classes.
Vocational and life skills programs can help prepare incarcerated individuals for life outside of correctional facilities. For instance, the Life Skills4 project is a combination of the Iowa Department of Corrections’ Going Home Re-Entry project and the Des Moines Area Community College’s Training Academy-Career Link. This program provides training on job readiness preparation, career development, vocational trades, and financial responsibility to enable those who are incarcerated to successfully reenter the community and obtain employment.
Creating more welcoming environments and providing accessible information about correctional facility policies and rules can also reduce the burden on families trying to connect children with incarcerated parents. For example, in Washington County (MN), the Jail Division has a Web page dedicated to visiting with children,5 which outlines visitation times and rules for visiting with minors. Creating more family-friendly visitor areas, such as the one in the Allegheny County (PA) Jail,6 can create a more welcoming environment for children in what can otherwise be an intimidating or tense environment.
Reducing the barriers to communication between incarcerated parents and children can support relationship development. These supports could include subsidizing or providing free phone calls to children, improving mailing practices to allow children to send their parents letters or drawings, and providing greater access to electronic communication such as emails or electronic messages (Peterson, Cramer, and Fontaine, 2019; Peterson et al., 2019). While incarcerated, parents can communicate with their children through phone calls, letter writing, or electronic communication.
Contact visits, when appropriate, can provide several benefits to both parents and children. They allow parents to practice parenting skills and can be “part of a larger family support program that may include parenting classes, coached telephone calls, case management, letter writing, and other activities” (Peterson, Cramer, and Fontaine, 2019; Peterson et al., 2019). Contact visits also enable children to better understand the incarcerated parent’s circumstances and build a relationship with their parent. Organizations like Community Works7 serve incarcerated parents to facilitate contact visits between parents and children.
3 See https://www.nh.gov/nhdoc/fcc/programs.html
4 See https://www.dmacc.edu/outreach/Pages/lifeskills.aspx
5 See https://www.co.washington.mn.us/3214/Inmate-Information
6 See https://www.alleghenycounty.us/jail/visitors/visitor-waiting-area.aspx
Enhancing Opportunities for Youth Learning and Development
Providing access to high-quality education and opportunities for learning and development is an important policy solution to explore with the goal of interrupting pathways that drive youth from racial minority groups into the criminal justice system. A well-established literature documents how disparities in access to quality education mirror disparities in the criminal justice system. These disparities are prevalent across key education indicators, including access to high-quality early childhood education (ECE) programs, school readiness, Head Start, and high school graduation (the National Academies, 2019a). These racial disparities in performance and educational attainment (e.g., grades, standardized-test scores, dropout rates) have been studied as an enduring achievement gap in education policy research.
This section explores programs, strategies, and models that can foster academic achievement and provide supports or buffers for students from racial and ethnic minority groups that can be protective from crime and criminal justice involvement. To this end, the committee reviews some of the literature on providing high-quality ECE and care, investigating promising charter school models, and reforming school disciplinary practices.
Early Childhood Education and Care
The ECE literature identifies high-quality early learning and care as protective against future involvement in criminal behavior and criminal justice involvement (Lochner and Moretti, 2004). ECE programs increase children’s cognitive, social, and health outcomes by enhancing children’s motivation for school and readiness to learn and identifying problems that impede learning (the National Academies, 2019a). Head Start, the public preschool program for children from disadvantaged backgrounds, is one example of an ECE program that has been associated with reduced likelihood of criminal justice involvement. After studying the long-term effects of Head Start, Garces and colleagues (2000) find that African American children who attended Head Start are 12 percentage points less likely to report ever being charged or convicted of any criminal offense when compared to siblings who did not attend. This relationship is even stronger among African Americans whose mothers’ level of educational attainment is high school completion. Head Start appears to have spillover effects as well; that is, individuals with an older sibling who attended Head Start are significantly less likely to report criminal activity as adults. Some research points to gender differences in treatment effects, with women showing greater reductions in crime than men after having participated in Head Start; conversely, men who participated tend to commit crimes that are more costly to society (García et al., 2018).
Some potential mechanisms that have been hypothesized to mediate the relationship between high-quality ECE and later criminal justice involvement are promotion of self-control, socio-emotional skills, reducing externalizing behaviors, and improved educational and labor market opportunities (García et al., 2019; Heckman et al., 2013). Not only do high-quality ECE programs reduce criminal justice involvement, but they also result in long-term societal cost-savings, which in large part can be attributed to crime reduction (García et al., 2020).
Charter School Models
Over the past 30 years, many states and school districts have expanded student access to charter schools. Charter schools are publicly funded, but they are not under the direct control of local school districts. They have additional freedom to experiment with educational programs and personnel policies.
In aggregate, the educational performance record of charter schools is not promising. Many appear to produce academic outcomes for students that are worse than the outcomes expected for similar students in traditional public schools. However, charter schools that serve disadvantaged urban communities have had noteworthy success, especially when these charters follow the “No Excuses” model. This model commits schools to additional time on task, extensive tutoring for students who are behind, the use of data to target students for additional help, the use of more intensive screening and accountability practices to hire and motivate educators, and cultural practices that communicate high expectations for all students. Neal (2018) summarizes a number of RCT evaluations of urban charters that follow this model. RCT evaluations are possible because these schools are often over-subscribed and districts use lotteries to ration admission.8
Studies conducted in Boston, Chicago, New York City, Texas, and Washington (DC) yield similar evaluation results for No Excuses charters, which all raise student achievement. The gains are typically larger in math than reading. Furthermore, more recent studies have shown that these schools increase future attendance and degree completion at four-year colleges. Dobbie and Fryer (2015) also examine nonacademic outcomes for students in New York City who applied to attend Promise Academy starting in middle school and find that attending Promise Academy reduced an index of risky behaviors among students in their late teens. They created the index using measures of teen pregnancy, criminal behavior, drug use, and
8 See Dobbie and Fryer, 2020, 2013, 2011; Hassrick et al., 2017; Curto and Fryer, 2014; Angrist et al., 2013; Abdulkadiroglu et al., 2011.
incarceration. Although these follow-up samples are small, the treatment versus control differences in risky behaviors are statistically significant.
To date, the direct evidence that attending No Excuses charter schools reduces criminal justice involvement for minority youth in disadvantaged urban communities is limited. The students in many of these studies have not been followed into their adult lives. However, these schools are often able to take urban students from disadvantaged backgrounds and improve their academic outcomes substantially. Many models of criminal behavior suggest that these improvements are likely to reduce criminal justice involvement.
As with any other successful intervention, it is not obvious how school systems can scale this model and maintain fidelity in implementation. Teachers in No Excuses schools work longer hours than teachers in traditional public schools, and the model places stringent performance demands on educators. It could be expensive to hire enough teachers who are both willing and able to implement this model faithfully on a much larger scale. Further, these schools do not work for all students, as some students do not thrive in high-expectation environments. However, they demonstrate that it is possible to give many students who live in our nation’s most challenging and distressed communities much better educations than they often receive from traditional public schools. Abdulkadiroglu and colleagues (2016) report that New Orleans (LA) had considerable success implementing No Excuses charters on a large scale as the city rebuilt its school system after Hurricane Katrina.
Another example of charter success is a network of schools operated by the University of Chicago. Hassrick and colleagues (2017) evaluate a cohort that applied for charter admission at (or near) the beginning grades of elementary school. These students had not already fallen below grade level nor were they switching from a public to a charter school to try to catch up. Therefore, the exceptionally long hours of individual tutoring that are so common in schools that follow the No Excuses model closely were not necessary. The vast majority of students who applied for admission to these schools came from Black families who were economically disadvantaged. Those who won lottery seats and entered these schools usually stayed. Attrition rates were low. Taken as a whole, the results suggest that attending these charters for grades K–5 raises a composite index of math and reading achievement by at least one half of a standard deviation, and Hassrick and colleagues (2017) find no evidence that these gains fade out as students move through higher grades.
It is difficult to compare test score impacts among studies that use different assessments, but the magnitude of the estimated learning gains produced by attending these University of Chicago charter schools is more than 50 percent of standard measures of the Black/White achievement
gap among elementary school students. The cohort of students who attended schools in this charter network can expect to have academic and labor market opportunities that will simply be unavailable to most of their peers in traditional public schools. Given that educational attainment and employment are strongly associated with the absence of a criminal record, there are good reasons to believe that these expanded opportunities may reduce their involvement in the criminal justice system as well.
Neal (2018) notes that the literature on both voucher schools and charter schools shows that, in general, private schools and charters are not clearly superior to traditional public schools. However, most studies that document important gains for students who gain access to new schooling options in the private or charter sectors are studies that document gains from disadvantaged students of color who live in urban areas with above-average crime rates.
In many cities, legislation or organized political opposition limit the expansion of the No Excuses model as well as other alternatives to traditional public schools. Policy makers need to prioritize the expansion of new and better educational opportunities for students in distressed urban Black, Latino, and Native American communities. However, policy makers must make sure that new schools that operate under alternative organizational models and governance structures remain accountable for student outcomes.
Reforming School Disciplinary Practices
Measures of school discipline reveal significant disparities by race. For example, in 2013, 15 percent of Black students received an out-of-school suspension, compared with four percent of White students and six percent of Latino students. Moreover, a disproportionate number of Native American youth, LGBQ/GNCT youth, and youth with disabilities are suspended or expelled from school as a result of discretionary disciplinary practices (American Bar Association, 2018; Poteat et al., 2016). Much of the existing research on the “school-to prison pipeline” suggests that the disparities in school discipline by race and ethnicity are associated with the disparities seen in juvenile justice involvement, including the fact that Black youth are more than twice as likely as White youth to be arrested as juveniles (5.4% compared to 2.1%).
Researchers have theorized multiple ways school suspension or expulsion could be correlated with an increase in juvenile arrest and detention. First, with the presence of school resource officers (SROs), any disciplinary infraction could increase the probability of interacting with the police force (Owens, 2017). Second, suspension reduces time spent in school and might increase the probability of arrest during the days of suspension (Mowen and
Brent, 2016). Third, suspension and expulsion reduce attachment to school, and this in turn could reduce the probability of high school graduation, which can increase the probability of future criminal activity (Lochner and Moretti, 2004). The existing school-to-prison pipeline research is primarily correlational and often does not examine behavioral characteristics among youth and their family backgrounds, which criminology research has shown to be powerful predictors of violence and system involvement (Graf et al., 2021; Labella and Masten, 2018).
Disparate treatment of students when it comes to disciplinary actions, such as suspension or expulsion, is well documented (see, e.g., Gordon, 2018). Most of the previous work has focused on racial/ethnic differences, finding that Black and Latino high school students, for example, are only slightly more likely than White or Asian students to be “sent to the principal’s office” for disciplinary infractions but are two to four times more likely to be suspended or expelled (Wallace et al., 2008). Interestingly, the reasons why White and minority students are reported for discipline differ significantly, with minority students being referred more often for more subjective reasons, such as “disrespect” and “perceived threat,” and White students being referred more often for more objectively identifiable reasons, such smoking or vandalism (Skiba et al., 2002). Other research documents the bias that young Black girls face, which translates into their being perceived by adults as less innocent and more adult-like than their White peers (Blake and Epstein, 2019; Epstein et al., 2017). The higher rates of suspension for Black and Latino students for subjective infractions is consistent with differences in teacher and school personnel treatment of students based on race and ethnicity.
Experimental evidence confirms racial bias in disciplinary decisions. Okonofua and Eberhardt (2015) find that when presented with identical descriptions of student behavior, teachers viewed two minimal infractions as more troubling and deserving of harsher punishment when committed by a Black student than when committed by a White student. Mendez (2003) finds that unequal treatment based on race, ethnicity, gender, sexuality, or disability status results in disparate outcomes for children, as out-of-school suspensions are highly predictive of future involvement with the criminal justice system and reduced educational achievement. Recent initiatives have sought to reduce the suspensions and expulsions of youth of color from schools, but to date these initiatives have been small in scope.
Given the disproportionate filtering of students into the juvenile justice system from schools, one strategy to reduce disparities in the juvenile justice system is to invest in alternative school disciplinary strategies, such as the use of multi-tiered systems of support (MTSS; Ricks and Esthappan, 2018; Solomon et al., 2018). Two promising MTSS models include Integrated School Supports and Positive Behavioral Intervention
Supports, both of which work with parents, schools, and other community stakeholders to address the specific needs of the students and schools they serve. Evaluations of these models conducted by the American Institutes for Research, MDRC, and ICF International have resulted in mixed reviews that range from null to positive, with no negative findings. While these strategies hold promise for ending the pathway from school discipline to justice system involvement and, ultimately, reducing the disproportionate number of youth of color coming into contact with the juvenile justice system, ongoing evaluations are needed to monitor program outcomes and implementation fidelity if these programs are to be successful in supporting adolescent development and academic success (Solomon et al., 2018).
Another program that has shown promise in reducing racial disparities in the juvenile justice system is the Philadelphia Police School Diversion Program. This program, which was started by the Philadelphia Police Department in 2014, allows students with no prior criminal history who have been arrested for misdemeanor crimes the chance to avoid being officially charged with a criminal offense. Rather than enter the justice system, these adolescents are moved into a diversion program designed to help them change their behaviors and life trajectory. A key component of the program is the assignment of a social worker and provision of academic support, mentoring, social and emotional competency building, and other services. Since the program began in 2014, there has been a 54 percent drop in arrests in the city’s schools. The program is currently participating in a three-year outcomes evaluation with funding provided by the Office of Juvenile Justice and Delinquency Prevention (Goldstein et al., 2019).
School Resource Officers
In addition to disciplinary practices, SRO programs have been subject to debate as they have been widely implemented and concerns about criminalizing student behavior have increased (Ryan et al., 2018). School security measures increased dramatically in response to high-profile incidents of lethal school violence (Theriot, 2009). One such measure has been the assignment of SROs to patrol schools and promote safety. Research in this area is limited, but there is evidence to suggest that the placement of SROs can criminalize youth, as it has been linked with increased arrests for non-criminal youth behavior and increased severity of punishment (King and Schindler, 2021; Weisburst, 2019; Na and Gottfredson, 2013). Gottfredson and colleagues (2020) examine effects of increased SROs in a sample of 33 public schools and find that the increase in SROs was associated with an increase in drug- and weapon-related offenses and
exclusionary disciplinary practices. On the other hand, Theriot (2009), reviewing arrest rates in 13 schools with SROs, finds that having an SRO did not predict more total arrests, and having an SRO placed in a school decreased the arrest rate of assault and weapons offenses.
Fisher and Hennessy (2016) conducted a systematic review and meta-analysis of SROs and exclusionary discipline in U.S. high schools. The authors found that the pattern of results across the separate random effects meta-analyses provides evidence that the presence of SROs in high schools is associated with higher levels of exclusionary discipline using 10 effect sizes from seven different studies. One of the two models run by the authors was statistically significant, indicating that the presence of SROs was associated with approximately one additional exclusionary discipline incident per week in a school of 1,500 students.
Since serious crime occurs rarely on school campuses, SROs spend most of their time investigating minor incidents (King and Schindler, 2021). Time spent investigating minor offenses creates an environment where “schools subject students to strict scrutiny” for behavior that would not reach this threshold had it occurred outside of campus (p. 30). One study found that as schools increase their use of police, they record more crimes involving weapons and drugs, and they report a higher percentage of their nonserious violent crimes to law enforcement (Na and Gottfredson, 2013).
For schools in under-resourced communities, resources directed toward SROs may be better directed to other social supports for students. A report by the Brookings-AEI Working Group on Criminal Justice Reform documents some of the negative consequences of SRO programs and offers the following strategies for reforming these programs to promote safety and to reduce harm.
Over the short term, the Brookings-AEI group recommends these reforms:
- Providing staff training and resources so students have support services; and
- Having school leadership strictly limit the roles and responsibilities of SROs.
Over the medium term, the group recommends these reforms:
- Eliminating funding for police in schools; and
- Removing police from schools to invest in services that are proven to improve safety.
Over the long term, the group recommends these reforms:
- Breaking the school-to-prison pipeline by implementing alternatives to suspensions and expulsions;
- Creating healthy school cultures based on developmentally appropriate behavioral approaches;
- Integrating school and community-based restorative justice practices instead of punitive measures; and
- Positively engaging families of students (King and Schindler, 2021).
An additional set of questions regarding SROs remain to be addressed by the evidence: what is the optimal role of SROs in schools given concerns about mass shootings in schools? If embedded in schools, how might this influence disciplinary and criminal justice outcomes for youth of color? How can schools promote safety and protect students of color from disparate treatment by SROs? These questions outline next steps for the research on school safety. Schools are an optimal setting for shaping healthy socioemotional development, prosocial relationships, and conflict resolution when they provide the right climate and support services for students. By disrupting pathways that lead to criminal justice involvement, such as expulsions and SRO referrals for non-criminal behaviors, schools can be a promising site of intervention for reducing racial disparities in criminal justice contact and arrests for youth.
Promoting Health and Well-being
The racial inequalities in crime and justice that are discussed throughout this report are also reflected in health and well-being indicators. Relatedly, the health effects of criminal justice contact and prolonged involvement are well documented (for example, see NRC, 2014).
Health Disparities in the Context of the Criminal Justice System
Numerous health disparities exist between persons involved in the criminal justice system and the general U.S. population (Bui et al., 2019). People in state or federal prisons are 1.5 times more likely than people in the general U.S. population to report ever having a chronic condition, and half of all people in state or federal prisons report having a chronic condition such as asthma, heart disease, or diabetes. Rates of infectious diseases, such as tuberculosis, HIV, hepatitis B and C, and sexually transmitted diseases are higher among the incarcerated population than among the general U.S. population (Maruschak et al., 2015). In 2005, more than half of the incarcerated population had a mental health problem, including 56 percent of state prisoners, 45 percent of federal prisoners, and 64 percent of jail inmates (James and Glaze, 2006). Moreover, 58 percent of state prisoners
and 63 percent of people sentenced in jail met the DSM (Diagnostic and Statistical Manual of Mental Disorders)-IV criteria for drug dependence or abuse during 2007–2009 (Bronson et al., 2017). Comparatively, 9.0 percent of persons ages 12 and older in the general U.S. population had a substance use disorder in 2009 (Substance Abuse and Mental Health Services Administration, 2015), and 11.3 percent of persons ages 18 and older had symptoms of serious psychological distress in 2005 (SAMHSA, 2006).
In addition to these disparities between the justice-involved population and the general U.S. population, health disparities among Black, Latino, and Native American populations that have been observed in the general U.S. population may also apply to the justice-involved population. These additional disparities may further compound the disadvantage of justice-involved Black, Latino, and Native populations compared with non-Hispanic White persons. With this complex relationship in mind, it is important to understand health disparities in the context of the criminal justice system and to explore potential interventions that mitigate the detrimental effects of criminal justice system involvement on health.
Health Care Policy and Crime
A small but important literature highlights a relationship between Medicaid expansion policies and crime. Aslim and colleagues (2022) found that increased health care access through Medicaid coverage expansion was associated with reduced recidivism among offenders who were convicted of violent and public order crimes. The authors postulate that an increase in referrals to addiction treatment with Medicaid coverage could reduce impulse behavior and, by extension, recidivism. Another study examines state-level crime data from 2009 to 2018 using a differences-in-differences design and reports that states that expanded Medicaid during this time period observed a 5.3 percent decrease in violent crime rates when compared to states that did not expand coverage (Vogler, 2020). The reduction in crime was also associated with an annual cost savings of approximately $4 billion. Moreover, there is evidence that Medicaid expansions were also associated with reductions in robbery, aggravated assault, and larceny theft, which have been theorized to be connected to increased substance use treatment. Simes and Jahn (2022) find evidence that supports this theory, with a differences-in-differences study that found that Medicaid expansion (2014–2016) was associated with 25–41 percent negative difference in drug arrests when compared to counties that did not expand Medicaid coverage. Additional research that explores the long-term effects of Medicaid expansion on crime as well as race-specific differences is needed in this area. Box 7-3 discusses health in the context of reentry and a community-based model for improving access to health care among justice-involved populations.
Promoting Mental Well-being for Youth
As mentioned above, criminal justice disparities also exist for people with mental illness. Among youth, those with serious mental illness are more than three times more likely to come into contact with the criminal justice system than their peers without mental illness (Erickson, 2012). Youth with diagnosed mental health disorders are overrepresented in the juvenile justice system (Meservey and Skowyra, 2015; Teplin et al., 2015; Shufelt and Cocozza, 2006), and this prevalence increases the further a youth is within the system (Wasserman et al., 2010). Incarceration has been shown to worsen the behaviors in youth that are associated with undiagnosed mental health illnesses (Erickson, 2012). Co-occurring conditions predict worse outcomes; for example, youth with co-occurring behavioral problems and emotional problems are at elevated risk for recidivism (Hoeve et al., 2013; Cottle et al., 2001) and committing violent offenses during young adulthood (Copeland et al., 2007). Thus, juvenile justice programs are responsible for the care of a large number of youth who have complex mental health needs (Cocozza and Skowyra, 2000).
Youth struggling with mental health challenges need effective treatment, such as wrap-around and diversion programs, as well as community-based mental health care specialized for youth. Research suggests that the presence of mental health care offices in a county can be protective in terms of reducing crime rates (Deza et al., 2022). There is little information on evidence-based practices and policies specifically for justice-involved transition-age youth with mental health problems (NRC, 2014; Hoffman et al., 2009). Most of what is known is extrapolated from studies with adult or adolescent justice-involved populations. A variety of treatments have been well validated to target delinquency among justice-involved adolescents (e.g., Multisystemic Therapy, Multidimensional Treatment Foster Care; for review, see Henggeler and Sheidow, 2012), but far fewer are specifically designed to meet the unique developmental needs of transition-age youth or to address mental health problems among justice-involved youth. Evidence-based treatments for adolescents can be adapted for use with transition-age youth; however, they often rely heavily on parental involvement, an approach that may be less effective or feasible with transition-age youth who may be living independently, living with peers, or otherwise disengaged from their family of origin.
In a study of collaboration between child welfare and juvenile justice, two factors predicted successful coordination of mental health services: (1) having a single agency held accountable for the youth’s well-being (i.e., either child welfare or juvenile justice) and (2) interagency sharing of administrative data (Chuang and Wells, 2010). Thus, effective coordination
of care and agency accountability are necessary to ensure that youth do not fall through the cracks.
Other research suggests that youth and young adults with a co-occurring serious mental illness and substance use disorder require treatment that addresses the “whole person.” Such person-centered services may include psychosocial interventions, family behavioral therapy, medication, proactive outreach, and use of specialized applications that can assist or provide an intervention and track symptoms (Brewer et al., 2017). In addition to targeted intervention for youth with mental illness, a more universal approach to promote well-being among youth who are exposed to violence is needed. As described above, living in areas characterized by high rates of violence is associated with poor mental health and well-being outcomes (e.g., anxiety, fear, posttraumatic stress disorder symptoms; Lee, 2016; Smith and Patton, 2016; Assari et al., 2015).
Diverting Youth with Mental Illness from the Justice System
Another promising strategy is the targeted diversion of youth with mental disorders to mental health treatment and other programs. Under mental health diversion programs, justice and social services agencies collaborate to divert youth offenders with mental disorders to mental health treatment in lieu of further court processing. It is hoped that—if mental health treatment is effective—diversion programs can help to reduce recidivism and the severity of crimes committed by offenders with mental disorders, thereby reducing the societal cost of crime.
One meta-analysis did not find significant reductions in recidivism, even for diversion programs that specifically targeted mental health needs (Schwalbe et al., 2012). However, when evidence-based interventions for adolescent delinquent behaviors (e.g., Multisystemic Therapy, Functional Family Therapy) were included in diversion plans, results were promising. Similar to these findings, preliminary results from Ohio’s Behavioral Health Juvenile Justice program suggest that a diversion program that provides evidence- and community-based behavioral health treatment is effective in improving both delinquency and behavioral health outcomes (Kretschmar et al., 2016). Thus, diversion programs may be effective when evidence-based treatments are available in youths’ own communities. Further, diversion programs reduce the time spent in locked settings, a contributor to developmental delays (Chung and Little, 2005). For these reasons, diversion programs need to be tailored to meet the needs of transition-age youth with mental health problems and examined as alternatives to formal sanctions.
Racial inequalities exist across the communities, institutions, and systems that shape and are shaped by the criminal justice system. As Chapter 3 asserts, structural policy reform that addresses the lasting effects of segregation, discriminatory housing policies, and other socioeconomic factors is needed to eliminate racial inequality. This chapter provides some promising strategies or policy solutions that, while situated outside of the criminal justice system, have the potential to mitigate the racial disparities that exist in crime, victimization, and criminal justice involvement. In summary, the committee offers the following conclusions that synthesize the evidence on promising approaches to address racial inequalities through non-criminal approaches.
CONCLUSION 7-1: Policies and programs that make progress toward eradicating disinvestment patterns show potential for reducing criminal justice system involvement and racial inequalities.
CONCLUSION 7-2: Improving the material well-being of people in disadvantaged neighborhoods (e.g., job opportunities, economic support) can reduce contact with the criminal justice system.
CONCLUSION 7-3: Ecological interventions that improve the quality of the built environment are a promising method of moderating the relationship between concentrated disadvantage and racially disparate outcomes in crime.
CONCLUSION 7-4: Given the enormous social cost and racial inequality associated with gun violence, measures for reducing the supply of guns, such as background checks and gun licensing, may contribute to reducing serious crime that is concentrated in Black, Latino, and Native American communities.
CONCLUSION 7-5: Evidence-based strategies in social policy institutions, such as education, health, and child welfare, to promote positive child and youth development and to promote the health and well-being of children and families are also needed to reduce racial disparities that exist in crime, victimization, and criminal justice involvement.
Addressing such “root causes” of racial inequalities is necessarily a long-term approach, one that will require commitment from a large swath of social institutions and stakeholders. Such approaches can be augmented by specific criminal justice system reforms that target reducing racial disparities at key points in the system. The committee turns to those strategies in the next chapter.