Prisoner Reentry and the Reproduction of Legal Cynicism

Abstract

More than 600,000 prisoners are released from incarceration each year in the United States, and most end up returning to metropolitan areas, concentrated in resource-deprived neighborhoods. To the extent that convicted criminals are distrustful of the criminal justice system, the funneling of massive numbers of former prisoners back into select neighborhoods likely facilitates the reproduction of legal cynicism in those areas. Accordingly, this study tests the effect of prisoner reentry on the culture of neighborhoods, particularly with regard to legal cynicism. Using two-waves of data on the geographic distribution of returning prisoners in Chicago from the Illinois Department of Corrections combined with data on neighborhood characteristics from the U.S. Census, the Chicago Police Department, the Project on Human Development in Chicago Neighborhoods, and the Chicago Community Adult Health Study, I conduct a cross-lagged analysis of the effect of the concentration of returning prisoners on legal cynicism as well as the effect of legal cynicism on the geographic distribution of returning prisoners. Findings reveal that a dense concentration of returning prisoners in a neighborhood facilitates the reproduction of cynical views of the law in the neighborhood. The substantial growth in the number of releases from prison and the stark concentration of the formerly incarcerated in select neighborhoods has detrimental consequences for the culture of receiving neighborhoods.

Más de 600.000 prisioneros son liberados de encarcelamientos cada año en los Estados Unidos, y la mayoría terminan regresando a las áreas metropolitanas, y se concentran en barrios privados de recursos. En la medida en que los criminales convictos desconfian del sistema de justicia criminal, la canalización de un número masivo de ex-prisioneros en barrios selectos probablemente facilita la reproducción de cinismo legal en esas áreas. Por consiguiente, este estudio pone a prueba el efecto de la reentrada de prisioneros en la cultura de los barrios, especialmente en relación con el cinismo legal. Utilizando informacion de la base de datos de la distribution geografica de pricioneros re-entrantes al sistema en Chicago del Departamento de Correcciones de Illinois en combinación con los datos sobre las características de barrios del censo de Estados Unidos, el Departamento de Policía de Chicago, el Proyecto de Desarrollo Humano en Barrios de Chicago, y el Estudio de la Salud de la Comunidad de Adultos en Chicago, se realiza un análisis denominado “cross-logged” sobre el efecto de la concentración de prisioneros re-entrantes al sistema en el cinismo legal, así como el efecto del cinismo legal sobre la distribución geográfica de los presos que regresan. Los hallazgos revelan que una densa concentración de retornos de los prisioneros a un barrio facilita la reproducción de visión cínica de la ley en los barrios de estos. El crecimiento sustancial en el número de liberaciones y la cruda concentración de ex-convictos en determinados barrios tiene consecuencias perjudiciales para la cultura de los barrios que los reciben.

prisoner reentry, legal cynicism, neighborhood, Project on Human Development in Chicago Neighborhoods (PHDCN), mass incarceration

One in 100 adults in the United States is currently in prison or jail, equaling over 2.2 million individuals (National Research Council 2014). A corollary statistic is the number of individuals exiting prison each year. In 1977, roughly 150,000 individuals were released from U.S. prisons. By 2007, yearly releases surpassed 720,000, representing a nearly 400 percent increase in three decades (Bureau of Justice Statistics 2000; West and Sabol 2008). Recently the number of yearly releases dropped below 700,000, but the volume of releases still surpasses 600,000 each year (Carson 2014). In total there are roughly five million formerly imprisoned individuals residing in U.S. neighborhoods (Shannon et al. 2015). Estimates suggest that up to half of prison releases are “churners” who have been in prison before (Langan and Levin 2002; see also Lynch and Sabol 2001).

Despite the sheer magnitude of returning prisoners in the United States, most neighborhoods are untouched by prisoner reentry. Prisoner reentry is highly concentrated in a relatively small number of neighborhoods, generally within metropolitan areas. For instance, research by the Urban Institute reveals that more than one-half of prisoners released from Illinois prisons in 2001 returned to Chicago, and one-third of these formerly incarcerated individuals were concentrated in only six community areas (La Vigne et al. 2003). These six communities were among the most economically and socially disadvantaged in the city.

The reasons why ex-prisoners tend to concentrate in the same neighborhoods include social ties to the neighborhood as well as the limited income, wealth, and job prospects of the typical offender. It is also the product of the unwillingness of owners and landlords in the private housing market to rent to felons and the combination of long waiting lists for public housing assistance and the unwillingness of public housing authorities to provide units or vouchers to formerly incarcerated individuals. Also of importance is the relative lack of affordable housing in the United States. Hence, former prisoners tend to cluster in those few neighborhoods where they are able to access and afford housing.

The consequences of prisoner reentry for receiving communities are beginning to become clear. In a study of Tallahassee neighborhoods, Todd Clear and associates (2003) find that the rate of prisoner reentry in a neighborhood in a given year is positively associated with neighborhood crime rates the next year (see also Dhondt 2012). Similarly, a study of census tracts in Seattle finds that the rates of prisoner reentry in neighborhoods are positively associated with subsequent violent crime rates (Drakulich et al. 2012). John Hipp and Daniel Yates (2009) find that an increase in the rate of parolees leads to an increase in various crime types (aggravated assault, robbery, and burglary).

Two recent studies examine the association between parolee concentration and recidivism. Using the property destruction and resulting dispersion of the parole population in Louisiana induced by Hurricane Katrina in 2005, Kirk (2015) finds that an increase in the concentration of parolees in a neighborhood leads to a significant increase in the reincarceration rate among former prisoners, whereas a dispersion of the parole population is associated with a decline in the reincarceration rate. Using data from Ohio, Alyssa Chamberlain and Danielle Wallace (forthcoming) similarly find a significant positive association between the geographic concentration of parolees and recidivism. Though recent research has established a significant, positive relationship between concentrated prisoner reentry and subsequent crime-related outcomes, there is a lack of empirical research on the mechanisms explaining the association.

Why does concentrated prisoner reentry appear to produce subsequent increases in neighborhood crime? One answer may be found in legal cynicism. Legal cynicism refers to a cultural orientation in which people perceive the law to be irrelevant to their everyday lives (Sampson 2012; Sampson and Bartusch 1998). In contrast to an attribute solely of individuals, Kirk and Andrew Papachristos (2011) suggest that legal cynicism can become part of the social fabric of neighborhoods: “Cynicism becomes cultural through social interaction. In this sense, individuals’ own experiential-based perception of the law becomes solidified through a collective process whereby residents develop a shared meaning of the behavior of the law” (p. 1201). Typically legal cynicism has been measured through survey responses to questions such as the following: “Laws are made to be broken” or “To make money, there are no right or wrong ways anymore, only easy ways and hard ways” (Sampson and Bartusch 1998:786). In turn, how people perceive the relevance of the law will affect their likelihood of obeying the law and their willingness to cooperate with the police to help locate a suspect or report a crime (Kirk et al. 2012; Kirk and Papachristos 2011; Sunshine and Tyler 2003). To the extent that convicted criminals are distrustful of the criminal justice system, the mass criminalization that has taken place in the last few decades may have had a devastating effect on perceptions of the law and authority across U.S. communities. The funneling of massive numbers of formerly incarcerated individuals back to select neighborhoods likely influences the social organization of the neighborhoods and reproduces a cultural ethos characterized by a severe cynicism of the law (Kirk and Papachristos 2011, 2015; Matsueda 2006; Sutherland 1947). Accordingly, in this article I ask: What are the effects of prisoner reentry on the culture of neighborhoods, particularly with regard to legal cynicism?

This article proceeds by first outlining mechanisms by which concentrated prisoner reentry may affect neighborhood crime. In this discussion, I expand on the theoretical link between prisoner reentry and neighborhood culture. I suggest that the churning of offenders between select metropolitan neighborhoods and prison works to reproduce a culture of legal cynicism in those neighborhoods. After outlining, theoretically, the mechanisms by which prisoner reentry leads to the perpetuation of neighborhood legal cynicism, I test this argument empirically by combining neighborhood-level data from the Illinois Department of Corrections, the U.S. Census, the Chicago Police Department, the Project on Human Development in Chicago Neighborhoods, and the Chicago Community Adult Health Study.

THE PRODUCTION OF LEGAL CYNICISM

Prior research on legal cynicism has demonstrated that cynical views of the law in a neighborhood are generally the product of the structural conditions of the neighborhood, particularly socioeconomic disadvantage, as well variation in police practices and procedural justice (Fagan and Tyler 2005; Kirk and Papachristos 2011; Sampson and Bartusch 1998). One key example is the work of Sampson and Bartusch (1998). Building on the work of Robert Kapsis (1978), Sampson and Bartusch theorize that the economic and political marginalization of neighborhoods, combined with racial segregation, breeds cynicism of the law in those neighborhoods.

Police practices, which are themselves influenced by neighborhood structural conditions (see, e.g., Kirk and Matsuda 2011), are also implicated in the production of legal cynicism. A robust body of research shows that compliance with the law is enhanced when neighborhood residents believe that laws are enforced fairly and when police adhere to procedural justice in their contacts with residents (Sunshine and Tyler 2003; Tyler 1990; Tyler and Fagan 2008). Procedural justice, in turn, promotes shared perceptions that the law and legal actors are legitimate. Conversely, adults and children alike report greater cynicism of the law when they view the actions of the police and other legal actors as unfair, unjust, and harsh (Fagan and Tyler 2005; Kirk et al. 2012). Whereas neighborhood structural conditions and the actions of legal actors are important sources of legal cynicism, the present article explores another likely source of this cynicism: the geographic concentration of prisoner reentry.

In the first articulation of the coercive mobility thesis, Dina Rose and Clear (1998) argue that high rates of incarceration in a neighborhood disrupt familial and neighborhood social networks, thereby undermining efforts to informally control neighborhood crime. This perspective recognizes that although criminal offenders often have a negative influence on communities, in many instances they are vital members of families and neighborhood social networks.

More than just the effect of incarceration, the coercive mobility thesis emphasizes the consequences of the cycling, or churning, of offenders in and out of a community. The return of a former prisoner to a neighborhood is preceded by the removal of the offender typically from the same (or proximate) neighborhood. Given the high rates of recidivism in the United States (Durose, Cooper, and Snyder 2014), the return of a prisoner to a neighborhood is often soon followed by another removal (Blumstein and Beck 2005; Clear, Waring, and Scully 2005; Lynch and Sabol 2001). This turnover in a neighborhood produced by the dual processes of prison admission and release can erode informal social control in the neighborhood.

Rose and Clear (1998; also Clear et al. 2003) also argue that the effect of the churning of individuals between prison and the community is nonlinear: at low levels of churning, existing mechanisms of informal social control can potentially counteract the instability on the neighborhood created by the churning. However, after a certain threshold in the spatial concentration of incarceration and prisoner reentry, the effect of further churning produces social disorganization and ultimately crime.

Most of the attention given to the coercive mobility thesis focuses on the implications for informal social control, but perhaps no less important are the consequences of concentrated prisoner reentry on perceptions of the law. One basis for this assertion is found in recent work by Vesla Weaver and Amy Lerman (2010). Their work reveals that direct experiences with incarceration or police harassment reduces an individual’s trust in the law and government. Involvement with the criminal justice system significantly depresses a person’s trust in government, with trust becoming increasingly damaged the more interaction an individual has with the criminal justice system (i.e., on average, imprisoned individuals will have less trust in the government than individuals who have only been stopped and questioned by the police). Arguably, then, ex-prisoners bring back to the neighborhood not only their experiences with the prison system, but also, more broadly, the cumulative negative experiences they have had with the police, the courts, and with prison.

Christopher Muller and Daniel Schrage’s (2014) research also supports the contention that concentrated prisoner reentry has negative consequences for perceptions of the law. Using the General Social Survey (GSS), they find that, in aggregate, mass incarceration has coincided with a declining trust in the government, and specifically the court system. Interestingly, findings diverge by race. African Americans’ beliefs about the fairness of the court system are relatively immune to rising rates of incarceration for African Americans, whereas for whites, trust in the courts declines with increasing rates of white incarceration. The authors suggest that perceptions among African Americans are less affected than perceptions among whites when incarceration rates increase, because of long-held suspicions about the government and the criminal justice system in the African American community. Muller and Schrage (2014) also analyze data from the Washington Post/Kaiser Family Foundation/Harvard University Survey of African American Men, and find that African Americans who have been incarcerated—or who have had a close friend or family member incarcerated—were much more likely to view the racial disproportionality in incarceration rates as a product of police and court bias than were African Americans who have not had any experience with incarceration.

Given that the formerly incarcerated and their family members and close friends are relatively more distrustful of the criminal justice system, concentrated prisoner reentry may have a devastating effect on perceptions of the law and authority in a community. Concentrating the formerly incarcerated in the same neighborhood saturates social networks with criminals and potentially leads to the contagious spread of legal cynicism. Papachristos, Tracey Meares, and Jeffrey Fagan (2012) find evidence of this contagion effect in their study of former offenders. The characteristics of one’s social networks do, they find, influence perceptions of legal cynicism: Having fewer criminal ties appears to reduce legal cynicism, and having a network saturated with other criminals or gang members increases legal cynicism. Furthermore, former offenders embedded in a network of noncriminals are less likely to engage in criminal activity than former offenders who are embedded in a social network with criminals.

Modes of Cultural Transmission

There are a variety of reasons why cynical views of the law may receive wider adoption than more trusting views in neighborhoods characterized by concentrated prisoner reentry. Luigi Luca Cavalli-Sforza and Marcus Feldman (1982) make a distinction between horizontal, vertical, and oblique modes of cultural transmission that helps to illustrate the ways legal cynicism may be transmitted between residents in a neighborhood. Horizontal transmission refers to the peer-to-peer transmission of cultural frames. For instance, in his theory of differential association, Edwin Sutherland (1947) posits that criminal behavior is learned through interaction in intimate social groups. The study by Papachristos and associates (2012) illustrates peer-to-peer transmission among former offenders; individuals are more likely to hold favorable views of the law if their social networks are not saturated with other criminals.

Vertical transmission of culture refers to transmission from parent to child (Cavalli-Sforza and Feldman 1982). Nearly two million children in the United States have a parent in prison, which equates to 1 in 43 children in the country (The Sentencing Project 2009). One in 15 black children has a parent in prison. The number of children with a parent in prison has roughly doubled in the past two decades, as has the number of mothers in prison. In total, more than 800,000 parents are incarcerated in federal and state prisons in the United States. The benefits of removing criminal parents from a home are widely debated, yet evidence strongly suggests that parental incarceration has serious and negative consequences for children in many aspects of health, well-being, and educational development (for a thorough analysis, see Wakefield and Wildeman 2014).

Parental incarceration can lead to the transmission of legal cynicism to children via multiple pathways. The return of a formerly incarcerated parent back into a household may directly expose children to cynical views of the justice system. Because incarceration has become so commonplace in the past few decades, hundreds of thousands of children face the return of formerly incarcerated parents each year, many of whom hold severely cynical views of the criminal justice system. Even prior to the return of a formerly incarcerated parent, youth may be exposed to cynical views of the law from the other parent. For instance, in her ethnography of female partners visiting male prisoners at San Quentin State Prison in California, Megan Comfort (2007) describes the perceived injustices inflicted upon the prisoners by the criminal justice system. Although some of the women Comfort interviewed recognized the guilt of their partner and viewed the punishment as deserved, the majority of respondents spoke of the criminal justice system as an unjust, oppressive institution; such perceptions of the system likely filter down to children. Indeed, research reveals that children may feel anxiety and anger against the justice system for the absence of a parent and the hardships experienced by the family as the result of an incarcerated parent (Braman 2004). In turn, children of incarcerated parents are far less likely to trust the government than children whose parents are not incarcerated, and children of the incarcerated are, therefore, less likely to be civically engaged (Lee, Porter, and Comfort 2014).

Finally, oblique transmission refers to transmission from non-parental members of one generation to members of a younger generation. Examples of oblique transmission include transmission from a teacher to students and from “old heads” in a neighborhood-to-neighborhood youth. For instance, as described by David Harding (2010), because of the threat of violent victimization, younger boys often seek out older neighborhood peers for protection, which may include individuals who have been incarcerated. This process, which Harding calls cross-cohort socialization, may ultimately lead to the oblique transmission of cultural models. Street gangs play a similar role in many neighborhoods as youth often turn to gangs for protection as well as socialization.

While even in the most disadvantaged neighborhoods many adult residents work or go to school, those older peers available in the neighborhood to socialize youths are often those residents with idle time, including the unemployed or individuals employed in the underground economy. Given the devastating effects of incarceration on employment prospects (Pager 2003; Western 2006), presumably the concentration of prison releases into select neighborhoods increases the number of unemployed or underemployed individuals occupying public spaces (Drakulich et al. 2012). Such individuals may transmit cultural frames that differ from the views of working or more engaged residents (Kirk and Papachristos 2011). Those individuals with the least amount of time and least likelihood of hanging out in public spaces are often the ones with more favorable views of the law. Thus, cynical views of the police and the law may gain more traction than positive views simply because of frequency. Furthermore, to the extent that a term of imprisonment enhances one’s “street cred,” at least with respect to knowledge about the criminal justice system, a former prisoner’s cynicism may carry more influence than the words of someone with little firsthand knowledge of the realities of punishment.¹

The Consequences of Legal Cynicism

Whereas the focus of this study is on examining the predictors of legal cynicism, it is pertinent to consider why cynicism of the law is relevant. As noted, mounting research reveals that cynicism of the law is positively related to neighborhood violence (Kirk and Papachristos 2011; Sampson, Morenoff, and Raudenbush 2005). One pathway by which legal cynicism fosters violence is by loosening the moral bind of the law (Kirk and Matsuda 2011). Accordingly, legal cynicism may be one mediating factor explaining the churning of offenders between neighborhoods and prison. I suggest that the return of prisoners to neighborhoods furthers the cynicism in those neighborhoods already marginalized by society. In turn, cynicism serves to loosen the moral bind of the law. When this occurs in neighborhoods lacking in community controls, crime and violence are more likely to proliferate. In this way, legal cynicism and mass incarceration—and the return of former prisoners that results from mass incarceration—are likely to be reciprocally correlated.

CURRENT STUDY

This study merges several literatures to investigate the effects of concentrated prisoner reentry on legal cynicism. It builds on work concerned with legal cynicism in urban sociology, criminology, and the psychology of law, which largely traces the source of legal cynicism to structural conditions of neighborhoods, particularly concentrated disadvantage and racial segregation, and to procedural injustices in the criminal justice system (Fagan and Tyler 2005; Kirk and Papachristos 2011; Sampson and Bartusch 1998; Tyler and Fagan 2008).

Using a unique combination of data on the geographic distribution of returning prisoners merged with longitudinal data on neighborhood cultural and social processes, I test the following hypothesis: Concentrated prisoner reentry has a positive, exponential relationship with the extent of legal cynicism in a neighborhood. Whereas legal cynicism may be minimally affected in neighborhoods with low rates of returning prisoners, neighborhoods with high concentrations of returning prisoners are likely to experience an exponential increase in cynicism of the law. Because I am testing whether there is a positive relationship between concentrated prisoner reentry and legal cynicism, I rely upon one-tailed tests of significance in analyses to follow.

DATA AND RESEARCH DESIGN

Data used in this study come from five sources: the Illinois Department of Corrections (IDOC), the Project on Human Development in Chicago Neighborhoods (PHDCN), the Chicago Community Adult Health Study (CCAHS), the U.S. Census, and the Chicago Police Department (CPD). The IDOC data were obtained from the Illinois Criminal Justice Information Authority and consist of information on the geographic distribution of prisoners released from prisons in Illinois from fiscal year 1996 to 2013.² Releases include those from new court commitments as well as rereleases from prison following a recommitment from a parole violation. I draw upon all 18 years of data for descriptive analyses, and then focus on the years of 1996 and 2002 specifically in the cross-lagged analysis described below. I use prison release data from 1996 and 2002 to coincide with the dates of data collection for the PHDCN survey (calendar year 1994-1995) and the CCAHS survey (2001-2003). There were 21,990 prison releases from adult IDOC facilities in 1996, and 37,707 releases in 2002 (representing a 71 percent increase in the number of prison releases).

The PHDCN is an interdisciplinary project that focuses on understanding the causes of juvenile delinquency, adult crime, and violence, among other outcomes. For the purposes of the PHDCN, neighborhood boundaries were operationally defined by combining all of the census tracts in Chicago into 343 neighborhood clusters, constructed to be “as ecologically meaningful as possible, composed of geographically contiguous census tracts, and internally homogeneous on key census indicators” (Sampson, Raudenbush, and Earls 1997:919). These census indicators include socioeconomic status, race/ethnicity, housing density, and family structure. An average of 8,000 residents comprises each of the 343 neighborhood clusters. The 1994-1995 Community Survey yielded a probability sample of 8,782 Chicago residents situated within the neighborhood clusters. My unit of analysis is the neighborhood cluster defined in the PHDCN data.

The CCAHS was designed to examine the neighborhood context of the health of residents, through the collection of survey data at the individual, household, and neighborhood levels (House et al. 2011). The CCAHS began as a collaboration with the PHDCN and shares many of the same sampling design characteristics as the PHDCN. In particular, neighborhood boundaries were operationally defined like those in the PHDCN. The 2001-2003 CCAHS survey, which is a modified version of the Community Survey used for the PHDCN, was administered to a sample of 3,105 adults across the 343 neighborhood clusters. Combining the PHDCN and CCAHS data allows for a two-wave, temporal assessment of changes in neighborhood conditions in Chicago.

Measures

One of my outcome variables is a measure of legal cynicism. The first wave (1994-1995) measure of legal cynicism is derived from the PHDCN Community Survey and the second wave (2001-2003) measure is taken from the CCAHS data collection. To measure legal cynicism at both time points, I combine the exact same four items from the respective surveys. Respondents of the surveys were asked the extent to which they agree to the following: (1) Laws are made to be broken; (2) It’s okay to do anything you want as long as you don’t hurt anyone; (3) To make money, there are no right or wrong ways anymore, only easy ways and hard ways; and (4) Nowadays a person has to live pretty much for today and let tomorrow take care of itself.³

Prior work by Sampson and Bartusch (1998) using the PHDCN data incorporated a fifth survey item in the legal cynicism measure: ”Fighting between friends or within family is nobody else’s business” (p. 786). While present in the PHDCN data, that item was not administered in the CCAHS survey.

I construct the legal cynicism scale for each time point via a multilevel item response model, with responses to each survey question nested within a respondent, and respondents nested within neighborhoods. From this item response model, I output a neighborhood specific empirical Bayes (EB) residual to use as my scale. These two scales represent the average levels of legal cynicism, in 1994-1995 and 2001-2003, respectively, across residents of each given neighborhood. Per my hypothesis related to the nonlinear relationship between concentrated prisoner reentry and legal cynicism, I compute the natural log of both legal cynicism measures, and use the natural logs in the cross-lagged path analysis described to follow.

The other outcome variable of the study is a measure of the concentration of released prisoners per 100 residents in a neighborhood. As noted previously, I use measures from both 1996 and 2002. Hence, the variable is measured as the number of individuals released to a given neighborhood in 1996 and 2002, respectively, divided by the total number of residents in the neighborhood in 1996 and 2002 and multiplied by 100.

The IDOC data made available for this project contain information on the number of former prisoners returning to a given ZIP code. Because the boundaries of Chicago ZIP codes do not align perfectly with the boundaries of the PHDCN neighborhood clusters, I apportion the number of returning prisoners to the PHDCN neighborhood cluster using tools available in ArcGIS based on overlap between a ZIP code’s geographic area and a given neighborhood cluster.

In terms of control variables, I utilize two measures of perceptions of the police from the PHDCN and CCAHS data collections. As noted earlier, unduly harsh and inequitable police practices may produce cynicism of the law. Moreover, perceptions of the police may be a confounding factor predictive of concentrated prisoner reentry as well.

Indicators of police perceptions differ across the PHDCN and CCAHS data collections. The former contains the following indicators of satisfaction with the police: (1) The police in this neighborhood are responsive to local issues; (2) The police are doing a good job in dealing with problems that really concern people in this neighborhood; (3) The police are not doing a good job in preventing crime in this neighborhood (reverse coded); (4) The police do a good job in responding to people in the neighborhood after they have been victims of crime; and (5) The police are not able to maintain order on the streets and sidewalks in the neighborhood (reverse coded). In contrast, the CCAHS data has available the following indicators of procedural justice: (1) The police are fair to all people regardless of their background; and (2) The police in your local community can be trusted. Similar to the method I used to construct the measures of legal cynicism, I construct these measures via multilevel item response models. For the sake of consistency, I generically label these measures “positive police perceptions.”

Neighborhood structural data come from the 1990 and 2000 U.S. Census. Consistent with the social disorganization thesis (e.g., Bursik and Grasmick 1993), I use four measures of neighborhood structure at both time points: (1) the proportion of foreign-born residents in the neighborhood, (2) population density, (3) concentrated disadvantage, and (4) residential stability. The proportion of foreign-born residents is a single-item variable, and population density is constructed by dividing the population size of a neighborhood by its land area (in miles).

Scales to measure concentrated disadvantage and residential stability for both 1990 and 2000 are created via principal components analysis. I pool data from both years into the same data set (i.e., each neighborhood had two observations). By doing so, the factor loadings for each of the census items do not vary across the two time points, thus ensuring comparability across time. The two resulting factors are based on the following eight items: (1) concentrated disadvantage: the percentages of families below the poverty line, of families receiving public assistance, of unemployed individuals in the civilian labor force, of female-headed families with children, of non-Hispanic black residents, and of residents under the age of 18; (2) residential stability: the percentage of residents five years old and older who lived in the same house five years earlier and the percentage of homes that are owner occupied.

Finally, under the assumption that neighborhood violence produces cynicism of the law (see, e.g., Kirk and Papachristos 2011; Sampson and Bartusch 1998) as well as impacts patterns of incarceration and prisoner reentry, I include measures of violence in the analyses. It is likely that neighborhood violence is correlated with, and proxies for, a more general level of crime in a neighborhood. Incident-level reported homicide data were obtained from the Chicago Police Department (CPD) for the years 1993 to 1995 as well as 1999 to 2001.⁴ Address information from each incident is used to geocode the location of the homicide to the corresponding PHDCN neighborhood cluster. Given that homicide is a rare event, it is a common practice to construct rates based on three-year averages (see, e.g., Morenoff, Sampson, and Raudenbush 2001). Accordingly, my statistical model includes a measure of the homicide rate from 1993 to 1995 as well as the rate from 1999 to 2001.

Analytical Framework

Analyses follow two paths. First, I provide a visual depiction of the trends in prisoner reentry in Illinois and the geography of prisoner reentry in Chicago. As part of this descriptive analysis, I assess the spatial clustering of prisoner reentry using both global and local measures of the Moran’s I statistic (Anselin 1995). The local Moran’s I statistic is an indicator of spatial association that can be used to measure the extent to which neighborhoods with large concentrations of returning prisoners cluster together in the same areas of the city and the extent to which neighborhoods with few returning prisoners cluster close to each other. It is computed as follows:

$$I_i=\frac{Z_i}{m_2}{\displaystyle \sum }_j{W}_{ij}{Z}_j,$$

where $m_2=\frac{{\displaystyle {\sum }_i{Z}_i^2}}{N}$⁠; Z_i and Z_j represent the deviations, for neighborhoods i and j, from the mean share of former prisoners for neighborhoods throughout the city; and W_ij is a matrix of spatial weights that designates a neighborhood j as contiguous with a focal neighborhood i if it is adjacent to the north, west, south, or east of the focal neighborhood. A high value of I_i indicates that a given neighborhood is clustered near other neighborhoods with high proportions of formerly incarcerated individuals.

The global Moran’s statistic can be used to assess the extent to which prisoner reentry is clustered or dispersed across the City of Chicago in aggregate, whereas the local statistic is useful for identifying where spatial clustering may be located in the city. The global statistic represents the sum of the local measures.

In one of the descriptive analyses to follow, I will examine the temporal trend in the global Moran’s I. A z-score can be used to assess the extent to which a given attribute, in this case the number of returning prisoners per 100 residents in a neighborhood, is spatially clustered. A statistically significant positive z-score would reveal that neighborhoods with high rates of returning prisoners tend to cluster together in the same parts of the city and that neighborhoods with relatively low rates of returning prisoners cluster together. A statistically significant negative z-score indicates dispersion—neighborhoods with high rates of returning prisoners are not located close to each other, and that neighborhoods with low rates of returning prisoners are found far away from other low rate neighborhoods. A non-significant z-score would suggest that the pattern of prisoner reentry is spatially random.

Upon establishing the extent to which there is evidence of the spatial clustering of prisoner reentry in Chicago as a whole, I will use the local measure of Moran’s I to visually examine the location of any clustering at three time points: 1996, 2002, and 2013. The first two time points closely coincide with the dates of data collection for the PHDCN survey (1994-1995) and the CCAHS survey (2001-2003), and the last time point is used to depict more recent patterns of spatial clustering in Chicago.

Following this descriptive analysis of the temporal and spatial patterns of prisoner reentry in Illinois, I turn to cross-lagged path models to examine the relationship between concentrated prisoner reentry and legal cynicism. I have already described the rationale underlying my hypothesis that the return of many former prisoners to a neighborhood reproduces a culture of legal cynicism in the neighborhood. Yet, it is likely the case that the relationship is reciprocal. Neighborhoods characterized by cynicism of the law are more likely to generate crime and the associated punishments such as incarceration. Neighborhoods with high rates of incarceration tend to be on the receiving end of high rates of returning prisoners. Hence, legal cynicism at one time point likely influences the subsequent concentration of returning prisoners.

I use three model fit indices to assess how well the model fits the data: chi-square, the comparative fit index (CFI) (Bentler 1990), and the root mean squared error of approximation (RMSEA). A CFI value above .90 indicates a reasonable fit to the data, and above .95 is a good fit. Similarly, a RMSEA value below .08 indicates a reasonable fit and a value below .05 indicates a close fit to the data (Acock 2013).

RESULTS

Figure 1 depicts the trends in prisoner release in Illinois, based on data from the National Prisoner Statistics Program collected by the Bureau of Justice Statistics (see Carson and Mulako-Wangota 2016). Figure 1 reveals almost continuous growth in the number of prisoners released each year from the late 1970s until 2006.⁵ Growth in prison releases accelerated in 1990 and again in 2001. From the late 1970s until the peak of releases in 2006, the number of releases increased 500 percent, well surpassing the percentage growth in releases nationwide. Trends in admissions follow a similar trajectory, which is indicative of the churning of offenders between prison and select urban neighborhoods. The implication is that the substantial rise in the churning of offenders in and out of Illinois neighborhoods may have had consequences for neighborhood culture.

Figure 1.

Prison Releases in Illinois, 1978−2013

Interestingly, this period of growth in the size of the reentry population has been marked by a suburbanization and exurbanization of prisoner reentry. As noted, research by the Urban Institute found that in 2001, roughly half of prison releases in Illinois returned to Chicago (La Vigne et al. 2003). As seen on the right axis in Figure 2, the percentage of prison releases returning to the City of Chicago declined incrementally between 1996 and 2005, and then declined dramatically between 2005 and 2009. By 2013, fewer than 40 percent of exiting prisoners in Illinois were returning to Chicago. This decline in the percent of Illinois prison releases returning specifically to the city limits of Chicago results from two trends (see Appendix Figure A1): the decline in the percent of Illinois prisoners returning to the wider Chicago-Naperville-Elgin core-based statistical area (CBSA) and the decline in the percent of returnees to the Chicago-Naperville-Elgin CBSA that reside in the Chicago city limits. That being said, the volume of prison releases to Chicago was roughly the same in 2013 as in 1996—about 12,000 prison releases—despite the relative shift of returning prisoners to areas outside of the Chicago city limits. This finding occurs because, as demonstrated in Figure 1, the total number of prison releases increased so dramatically over the time period.

Figure 2.

Number and Percent of Illinois Prison Releases Returning to the City of Chicago, 1996−2013

Coinciding with this geographic shift in prisoner reentry is a decline, albeit a bumpy one, in the spatial clustering of prisoner reentry in Chicago. Figure 3 displays the trends in a global measure of spatial autocorrelation, Moran’s I. Comparing the z-scores over time provides an indication of whether spatial clustering becomes more or less intense. In all years, we see a z-score well above 3.09 (i.e., where the p value = .001), indicating that prisoner reentry is far more spatially clustered than would be expected if underlying spatial processes were random. However, the z-scores fluctuate. Between 1996 and 1999, the overall spatial clustering of prisoner reentry declines in Chicago. Between 1999 and 2003, it increases a substantial amount. During this period, the number of releases to Chicago increased from 12,288 to 17,253, yet the increase appeared to be spatially concentrated. Starting in 2003, the spatial clustering declines almost continuously through 2013, with the exception being an uptick in 2008.

Figure 3.

Z-Score for Global Moran’s I Measure of the Spatial Clustering of Returning Prisoners in Chicago Neighborhoods

The global measure of Moran’s I, which represents the sum of the local Moran’s I for each neighborhood in Chicago, reveals a downward trend in the spatial clustering of prisoner reentry in Chicago. If concentrated prisoner reentry is positively associated with legal cynicism in a neighborhood, as I hypothesize, then this suggests that the decline in spatial clustering should attenuate some of the cynicism of the law found in Chicago neighborhoods. Before testing my hypothesis, I continue the descriptive analysis by highlighting where the spatial clustering of prisoner reentry is located in Chicago.

Figure 4 maps the local Moran’s I, specifically highlighting those sections of the city where there are neighborhoods with high rates of prisoner reentry clustered close together (i.e., “high-high”), as well as other combinations of clustering (high-low, low-high, and low-low). Former prisoners are not evenly spread across Chicago. In each year the geographic distribution of returning prisoners is highly concentrated. In 1996, the clustering of former prisoners was mostly isolated to the neighborhoods west of downtown. These community areas include: Austin, West Garfield Park, East Garfield Park, Humboldt Park, North Lawndale, and South Lawndale. In 2002, we see the same spatial clustering west of downtown, but we also see a clustering on the South Side of Chicago, in community areas such as West Englewood, Englewood, and Grand Crossing, as well as on the southern border of the city in Riverdale. Moving to 2013, which Figure 3 indicated was the year with the lowest level of aggregate spatial clustering between 1996 and 2013, we see that the spatial clustering west of downtown Chicago has dispersed to some extent and pushed east. We also see that the spatial clustering in the middle of the South Side has dissipated, and that the clustering near the southern border of the city has grown to include the community area of South Deering. Overall, in 1996, 9 percent of Chicago neighborhoods were neighborhoods with high rates of prisoner reentry located adjacent to other high rate neighborhoods. In 2002, it had increased to 14 percent of neighborhoods. In 2013, it declined to 8 percent of neighborhoods. More interesting is the stability and instability in the location of returning prisoners. The west side of Chicago has and continues to be the place of residence for significant clusters of former prisoners, but with some dispersion between 2002 and 2013. The patterns of prisoner reentry on the South Side changed considerably over the course of the 1996-2013 time period. This is likely due, in part, to gentrification and the demolition of thousands of units of high-rise public housing such as the Robert Taylor Homes in and around the Englewood community area.

Figure 4.

Local Moran’s I Measure of the Spatial Clustering of Returning Prisoners in Chicago Neighborhoods across Three Points: 1996, 2002, and 2013

Table 1 displays a descriptive summary of the social, cultural, and sociodemographic characteristics of Chicago neighborhoods. Several interesting findings emerge from this table. As expected, I find a significant, positive relationship between legal cynicism and concentrated prisoner reentry. Legal cynicism is also significantly and positively correlated with concentrated disadvantage and rates of homicide, and negatively correlated with positive police perceptions. Concentrated prisoner reentry (the second set of variables) is common in neighborhoods characterized by high levels of disadvantage and homicide, and it is uncommon in immigrant neighborhoods, in areas with low population density, and in neighborhoods characterized by positive perceptions of the police. The sizable correlation between the measures of homicide and concentrated disadvantage reveals that neighborhoods in Chicago with the highest rates of homicide tend to be characterized by the most severe levels of socioeconomic disadvantage.

Table 1.

Correlations among Neighborhood-Level Variables, Chicago Neighborhood Clusters

		1a.	1b.	2a.	2b.	3a.	3b.	4a.	4b.	5a.	5b.	6a.	6b.	7a.	7b.	8a.	8b.
1a.	Legal cynicism, 2001-2003	–
1b.	Legal cynicism, 1994-1995	.43***	–
2a.	Concentrated prisoner reentry, 2002	.24***	.33***	–
2b.	Concentrated prisoner reentry, 1996	.22***	.33***	.92***	–
3a.	Positive police perceptions, 2001-2003	−.24***	−.47***	−.37***	−.34***	–
3b.	Positive police perceptions, 1994-1995	−.42***	−.79***	−.33***	−.33***	.49***	–
4a.	Concentrated disadvantage, 2000	.23***	.60***	.49***	.43***	−.64***	−.64***	–
4b.	Concentrated disadvantage, 1990	.25***	.61***	.50***	.46***	−.61***	−.67***	.95***	–
5a.	Residential stability, 2000	.10	−.02	.06	.04	.12*	.15**	−.21***	−.24***	–
5b.	Residential stability, 1990	.08	−.05	.04	.04	.15**	.18***	−.23***	−.26***	.93***	–
6a.	Foreign born, 2000	−.03	−.18***	−.43***	−.34***	.40***	.14*	−.63***	−.59***	−.18***	−.19***	–
6b.	Foreign born, 1990	−.02	−.14*	−.41***	−.32***	.34***	.07	−.56***	−.52***	−.27***	−.29***	.93***	–
7a.	Population density, 2000	−.01	.03	−.34***	−.34***	.00	−.21***	−.06	−.05	−.52***	−.56***	.42***	.46***	–
7b.	Population density, 1990	.01	.14*	−.28***	−.29***	−.10	−.31***	.11*	.12*	−.54***	−.56***	.27***	.34***	.96***	–
8a.	Homicide rate, 1999-2001	.30***	.62***	.54***	.51***	−.53***	−.59***	.73***	.75***	.01	−.02	−.43***	−.40***	−.12*	.02	–
8b.	Homicide rate, 1993-1995	.32***	.57***	.51***	.50***	−.55***	−.61***	.71***	.69***	.01	−.01	−.49***	−.43***	−.14*	−.02	.71***	–

Notes: N = 342 (excludes O'Hare Airport)

^*p < .05 ** p < .01 *** p < .001 (two-tailed tests)

In terms of the stability of legal cynicism and concentrated prisoner reentry across time, I find a correlation of .43 between the two measurements of legal cynicism. While highly significant, the rank ordering of neighborhoods by legal cynicism did change a considerable amount across time points. However, with concentrated prisoner reentry, the correlation across time points equals .92. Those neighborhoods with the greatest concentrations of prison releases tended to be the same across time points, even as the volume of prison releases (as seen in Figure 1) and the spatial clustering (Figure 3) increased dramatically between 1996 and 2002.

Figure 5 presents the conceptual model of the cross-lagged path analysis I estimate to examine the relationship between concentrated prisoner reentry and legal cynicism. In this model, various exogenous features of neighborhoods (i.e., the time 1 and time 2 covariates) are used to predict the concentration of prisoner reentry at a given time point, and also the extent of neighborhood legal cynicism. These predictors include perceptions of the police, concentrated disadvantage, residential stability, the homicide rate, population density, and the proportion of foreign-born residents in the neighborhood. Paths a and b represent the within-time point correlations between concentrated prisoner reentry and legal cynicism. Paths c and d represent stability coefficients for prisoner reentry and legal cynicism, respectively. These paths reflect the stability in prisoner reentry and legal cynicism over the course of the two waves of data. Including these stability coefficients allows me to account for prior variation in my two outcomes when estimating the reciprocal effects of concentrated prisoner reentry and legal cynicism.

Figure 5.

Cross-Lagged Path Model: Reciprocal Effects Between Concentrated Prisoner Reentry and Legal Cynicism

The paths e and f are the primary interest of the study and will reveal the association between concentrated prisoner reentry and subsequent legal cynicism as well as the association between legal cynicism and the subsequent geographic distribution of returning prisoners, net of the lags of these measures and other covariates. This conceptual model is implemented using the structural equation modeling (SEM) builder in Stata.

Table 2 presents results from the cross-lagged estimates of the conceptual model depicted in Figure 5. The model fit indices shown in Table 2 reveal that the model is a reasonable fit to the data. Whereas the statistically significant χ² value indicates that my predicted covariance matrix does not perfectly reproduce the observed covariance matrix, χ² values are sensitive to sample size, particularly when the sample size exceeds 200 as in the case here. For this reason, an alternative measure of fit using χ² is the ratio of χ² to the degrees of freedom. A small χ² value relative to degrees of freedom is indicative of good model fit. A ratio of 3 or below is generally regarded as indication of reasonable fit to the data (Kline 1998). In the case here, the ratio equals 2.82 (67.75/24), indicating good fit. Similarly, both the CFI (.967) and RMSEA (.073) demonstrate that the model fits the data well.⁶

Table 2.

Cross-Lagged Path Models of Concentrated Prisoner Reentry and Legal Cyncism

	b	(SE)	β
Natural log of legal cynicism, 1994-1995 (time 1)
Positive police perceptions, 1994-1995	−.102	(.013)***	−.532
Concentrated disadvantage, 1990	.019	(.004)***	.363
Residential stability, 1990	.005	(.002)**	.096
Foreign born, 1990	.049	(.016)**	.141
Population density, 1990	−.002	(.002)	−.044
Homicide rate, 1993-1995	.002	(.003)	.035
Constant	1.176	(.037)***
Concentrated prisoner reentry, 1996 (time 1)
Positive police perceptions, 1994-1995	−.183	(.199)	−.092
Concentrated disadvantage, 1990	.134	(.062)*	.254
Residential stability, 1990	−.082	(.029)**	−.151
Foreign born, 1990	.079	(.232)	.022
Population density, 1990	−.216	(.048)***	−.417
Homicide rate, 1993-1995	.145	(.063)*	.254
Constant	1.174	(.623)*
Natural log of legal cynicism, 2001-2003 (time 2)
Positive police perceptions, 2001-2003	−.005	(.005)	−.056
Concentrated disadvantage, 2000	.009	(.004)*	.207
Residential stability, 2000	.006	(.002)**	.154
Foreign born, 2000	.042	(.015)**	.195
Population density, 2000	.001	(.002)	.036
Homicide rate, 1999-2001	−.003	(.011)	−.023
LN(legal cynicism), 1994-1995	.200	(.057)***	.278
Concentrated prisoner reentry, 1996	.007	(.004)*	.096
Constant	.593	(.051)
Concentrated prisoner reentry, 2002 (time 2)
Positive police perceptions, 2001-2003	−.012	(.056)	−.006
Concentrated disadvantage, 2000	.085	(.047)*	.083
Residential stability, 2000	.011	(.027)	.012
Foreign born, 2000	−.430	(.105)***	−.080
Population density, 2000	.004	(.022)	.004
Homicide rate, 1999-2001	.105	(.147)	.031
LN(legal cynicism), 1994-1995	−.752	(.530)	−.042
Concentrated prisoner reentry, 1996	1.472	(.114)***	.852
Constant	.847	(.460)*
Within-time paths
Concentrated prisoner reentryt1 ↔ LN(legal cynicism)t1	.000	(.001)	−.023
Concentrated prisoner reentryt2 ↔ LN(legal cynicism)t2	.001	(.001)*	.091
Model fit
χ2(df)	67.75 (24)
CFI	.967
RMSEA	.073

Notes: N = 342 (excludes O'Hare Airport). Coefficients and standard errors have been divided by 10,000 for measures of population density and by 100 for measures of homicide. CFI = Comparative Fit Index. RMSEA = root mean square error of approximation.

^*p < .05 ** p < .01 *** p < .001 (one-tailed tests)

Figure 6 visually displays the standardized coefficients for the within-time correlations (i.e., paths a and b), autoregressive paths (c and d), and cross-lagged paths (e and f) found in Table 2. The within-time correlation at time 1 (path a) reveals little concurrent relationship between concentrated prisoner reentry and legal cynicism after controlling for other predictors in the model. However, I do find a significant within-time correlation at time 2 (path b). Moreover, I may also find a cross-lagged effect, if the consequences of prisoner reentry and legal cynicism for neighborhoods unfold over time.

Figure 6.

Cross-Lagged Path Model Results: Concentrated Prisoner Reentry and Legal Cynicism

Notes: Solid paths are statistically significant and dotted paths are non-significant. Standardized coefficients are shown for significant paths.* p < .05 ** p < .01 *** p < .001 (one-tailed tests)

The autoregressive path c from concentrated prisoner reentry at time 1 to time 2 reveals a very high degree of stability in inter-neighborhood variation in the concentration of prisoner reentry (standardized coefficient = .852). Despite a fair degree of change in the geographic patterns of prisoner reentry in Illinois, as indicated in Figures 2 and 3 (i.e., the increasing number of returning prisoners and spatial clustering between 1996 and 2002), the rank ordering of neighborhoods by the rate of returning prisoners per 100 residents in the neighborhood remained quite stable.

The autoregressive path d from legal cynicism at time 1 to time 2 is highly significant; however, the magnitude of the relationship is far lower relative to the autoregressive path between the two measures of concentrated prisoner reentry (standardized coefficient = .278). Whereas past levels of legal cynicism in a neighborhood are an important determinant of future levels, the relatively small standardized coefficient reveals a fair amount of change in the rank ordering of Chicago neighborhoods by legal cynicism. This finding, however, is unsurprising given that the economic and racial stratification of neighborhoods is a key predictor of legal cynicism (Kirk and Papachristos 2011; Sampson and Bartusch 1998), and that the period from 1996 to 2002 was marked by gentrification in several sections of Chicago as well as the demolition of some of the most infamous public housing sites in the country (e.g., Cabrini-Green and the Robert Taylor Homes).

The cross-lagged path e reveals a significant positive association between concentrated prisoner reentry at time 1 and legal cynicism at time 2. The standardized coefficient equals .096. Comparatively, as revealed in Table 2, other neighborhood factors such as concentrated disadvantage have a stronger association with legal cynicism than does concentrated prisoner reentry (standardized coefficient = .207). Nevertheless, in support of my main hypothesis, a dense concentration of returning prisoners in a neighborhood facilitates the production of cynical views of the law in the neighborhood.

I do not find evidence of a reciprocal path between legal cynicism at time 1 and concentrated prisoner reentry at time 2. The cross-lagged path f is not statistically different from zero. As shown in Table 2, what appears to influence the geographic distribution of returning prisoners in 2002 is—in addition to prior rates of returning prisoners—concentrated disadvantage and the proportion of foreign-born residents in the neighborhood.

I noted in footnote 6 that the model with the log-transformed measures of legal cynicism provides a slightly better fit than the model with the untransformed measures. Figure 7 visually depicts the expected relationship between concentrated prisoner reentry and legal cynicism. Whereas a model specifying an exponential relationship between concentrated prisoner reentry and legal cynicism fits better than specifying a linear relationship, the relationship visually appears to be nearly linear. Nevertheless, legal cynicism increases as the concentration of returning prisoners increases.

Figure 7.

The Expected Relationship between Concentrated Prisoner Reentry and Neighborhood Cynicism of the Law

DISCUSSION

In a recent article, Benjamin Justice and Meares (2014) suggest, “Prison life is characterized by treatment of inmates in discord with the principles of procedural justice” (p. 170). It is little surprise, then, that experiences with incarceration, and criminal sanctions more generally, tend to reduce an individual’s trust in the law (Muller and Schrage 2014; Weaver and Lerman 2010). The present article has sought to examine the implications of the return of dense concentrations of former prisoners for the culture of neighborhoods. I asked: what are the consequences of prisoner reentry for cynicism of the law among neighborhood residents? I find that concentrated prisoner reentry in a neighborhood is positively associated with subsequent levels of neighborhood legal cynicism, net of other neighborhood characteristics such as economic disadvantage and violence as well as a lagged measure of legal cynicism.

As noted, the reasons why ex-prisoners tend to concentrate in the same neighborhoods include personal factors such as social ties to the neighborhood. Indeed, the vast majority of newly released prisoners report residing with family members upon their exit from prison (Roman and Travis 2004). Of course residence with family members may be the result of more than preference; it may also result because of limited options for other housing. The lack of housing for ex-offenders is certainly a function of the limited income, wealth, and job prospects of the typical offender, and it is also the product of the unwillingness of owners and landlords in the private housing market to rent to felons and the combination of long waiting lists for public housing assistance and subsidies and the unwillingness of public housing authorities to provide units or vouchers to formerly incarcerated individuals. Also of importance is the fact that the availability of affordable housing in the United States has shrunk over the past decade at the same time that the number of households with extremely low incomes has drastically increased. For instance, the Joint Center for Housing Studies of Harvard University (2013) reports that the share of severely burdened renter households in the United States, defined as households paying more than half of their income for housing, increased from 20 percent to 28 percent from 2001 to 2011. For all these reasons, individuals exiting prison tend to cluster into the same few resource-deprived neighborhoods. Traditionally this clustering of former prisoners occurred in inner cities. However, increasingly former prisoners are locating outside of inner-city neighborhoods (see Figure 2), and this geographic shift coincides with the increasing suburbanization of poverty (see, for example, Kneebone and Berube 2013). That being said, whether in the core of central cities or in suburban areas, prisoner reentry continues to be highly concentrated for the reasons just described. Results of the present study suggest that this clustering facilitates the contagious spread of legal cynicism among residents of a neighborhood.

Per the preceding discussion, three limitations of the current study include the time period of the available data, the focus on urban neighborhoods to the neglect of the wider metropolitan area, and the lack of data on prison admissions to go along with the data on prison releases. I relied upon prisoner reentry data from the mid-1990s and early-2000s in order to align with the availability of data on neighborhood characteristics from the PHDCN and CCAHS. Other than the L.A. FANS data, with two waves of neighborhood data from 2000-2001 and 2006-2008, I am unaware of any data from the last few years providing as extensive a measurement of neighborhood conditions and social processes at multiple time points as the PHDCN and CCAHS. It would, nevertheless, be informative to examine the associations between concentrated prisoner reentry and neighborhood conditions using more recent data than employed in the present study. In terms of other avenues of future research, the focus in this study has been on the urban environment, but prisoner reentry is increasingly a challenge to entire metropolitan regions. Hence, a second avenue of future research is to examine the implications of the suburbanization of prisoner reentry on the culture and crime rates of suburban neighborhoods. A third avenue of research is to investigate the effect of the churning of offenders between prison and neighborhoods rather than just focus on the release of prisoners. I suggested in my discussion of the vertical transmission of legal cynicism that the removal of an incarcerated parent from a home or family can be detrimental to the views of authority figures among the children. More generally, a key assertion of the coercive mobility thesis (Clear et al. 2005) is that it is the churning of offenders that is destructive for informal social control and not simply the return of former prisoners. Hence, the present analysis should be extended by examining the effect of the combination of prison removals and returns.

In terms of implications, because research reveals that legal cynicism is positively predictive of violence (Kirk and Papachristos 2011; Sampson et al. 2005), and because violence remains stubbornly high in many neighborhoods despite a two-decade decline in reported rates of violent crime, it is imperative to contemplate how to restore positive perceptions of the law and the criminal justice system. One solution that follows from the results of this study is to disperse the formerly incarcerated population instead of concentrating it into select urban neighborhoods. This could be achieved through the creation of affordable housing options outside of disadvantaged areas. One way to promote this would be to provide tax credits or other financial incentives to real estate developers to provide affordable housing, with an agreement that individuals with criminal records not be barred from the housing.

Another potential solution is to lessen the use of incarceration in the United States and, more generally, to find an alternative to the culture of control that has characterized the U.S criminal justice system since at least the 1970s. A development in this regard is the use of alternatives to prison for low-level offenders, exemplified by California’s experiment in “realignment.” The Criminal Justice Realignment Act of 2011 shifts the responsibility for managing low-level felons from the state to the counties and is arguably the most consequential penal experiment since mass incarceration began. Joan Petersilia and Jessica Snyder (2013) estimate that in 2012 and 2013 alone, nearly 100,000 offenders who would have been incarcerated in a state prison in the previous era saw their punishments managed by counties instead. Time will tell if the declining use of state imprisonment in California leads to reductions in cynicism of the law and crime.

While solutions to minimize the detrimental effects of prisoner reentry on neighborhood culture and ultimately neighborhood violence are easy to conceive, implementing the necessary changes is anything but straightforward. The hyperincarceration of so many individuals over the past four decades likely means that legal cynicism will prevail in many communities for some time.

1

Similarly, Sutherland (1947) argued that the adoption of definitions favorable to the violation of the law depends upon the “intensity” of the definitions, which is a function of the intensity of the relationship between the sender and receiver of the definitions as well as the prestige of the sender. As a contemporary example, the acclaimed Chicago Ceasefire program harnesses the legitimacy and credibility of former offenders, the so-called violence interrupters, in order diffuse violent situations. Nevertheless, the mechanism is similar to the transmission of legal cynicism—the words of former offenders carry significant weight in certain groups.

2

The fiscal year in Illinois starts on July 1st and includes the last six months of the prior calendar year through the first six months of the next calendar year (e.g., FY1996 ran from 7/1/1995 – 6/30/1996).

3

Measurement of “legal cynicism” varies to some degree in the research literature. Sampson and Bartusch (1998), Sampson and colleagues (2005), Fagan and Tyler (2005), and Sampson (2012) use measures that tap a broad construct that combines cynicism of the law with moral cynicism and anomie, whereas Kirk and Papachristos (2011) more narrowly focus on perceptions of the “legal” system and the police in particular. The measure I use here adheres to the broader focus on legal and moral cynicism.

4

Data were provided by CPD’s Division of Research and Development. Findings from use of these data in no way represent the views of CPD or the City of Chicago.

5

Releases include both conditional (e.g., parole) and unconditional (i.e., expiration of the prison term) releases, but exclude prison exits from death, escape, and other infrequent reasons.

6

I compared these model fit indices, which are from a model with the natural logarithm of legal cynicism as outcome measures, with a model with the untransformed measures of legal cynicism. Model fit to the data is slightly better with the log transformed measures. For the untransformed data, the ratio of χ² to the degrees of freedom = 2.94, CFI = .965, and RMSEA = .075.

ACKNOWLEDGEMENTS

The author wishes to thank Natasha Frost and Mary Rose for helpful comments on an earlier version of this article. The author is grateful to Carmen Gutierrez for research assistance.

APPENDIX

Figure A1.

Percent of Illinois Prisoners Returning to the Chicago-Naperville-Elgin CBSA