Criminological Highlights Vol. 21, No. 2 - December 2023

This issue of Criminological Highlights addresses the following questions:

1. How did American news organizations contribute to American mass incarceration?
2. How do many police departments see their relationships to ordinary citizens?
3. Should communities welcome the implementation of “school resource officers” [police attached to ordinary schools]?
4. Is the impact of school suspensions the same across racialized groups of students?
5. How do ordinary political affiliations affect ordinary policing?
6. Why do so many people who don’t have stable housing have criminal records?
7. Are laws prohibiting employers from asking about the criminal records of job applicants effective in helping this group get jobs?
8. Why is it important, when attempting to control incarceration rates, to look beyond the laws related to sentencing and imprisonment? 

Item 1

The timing and persistence of prison growth in the United States since the 1970s can be understood by looking at the manner in which media outlets in the US circulated narratives portraying Blacks as being responsible for violent crime. The media effect itself was intimately intertwined with changes in the US homicide rate.

One of history’s most dramatic criminal justice changes is the increase in imprisonment rates that took place in the US between the mid-1970s and the first decade of the 21^st century. This paper examines the hypothesis that this growth was related to racial threat: “that the criminal justice system operates to suppress the competitive power of large minority groups” (p. 561) – in this case Black Americans.

Media stories present threat narratives “that define which groups are threatening and the type of threat posed…. [These] shape threat perceptions, exert pressure on institutional and political actors, and adapt to [changes such as] sustained increase in serious offending…  And, when media discourse portrays racial minorities as posing a criminal threat, social control increases as well” (p. 562). Criminal threat narratives can lead majority groups to support imprisonment and can “increase lawmakers’ willingness to pursue punitive policies… [designed to] align with the images of public opinion and crime that circulate media discourse” (p. 565). We have seen similar findings with respect to fear of crime (Criminological Highlights 20(1)#2).

Unlike many papers that examine only the period during which US imprisonment rates increased dramatically, this paper examines a much longer period (1926-2016) in order to understand the role of crime and threat in quite different historical periods. It examines over one million news articles published in 4,177 outlets between 1926 and 2016. These stories related to crime, employment, voting and race.

The articles were coded to determine whether Black people were portrayed “as holding substantial economic or political influence or [were described as] being associated with crime… in a negative, fearful or hostile light” (p. 568).  Because of the size of the task, a subset of the articles was first coded by three people (who showed very high agreement in their coding). Subsequently, algorithmic processing of the full data set was carried out after determining that the human and machine coders showed very high levels of agreement.

The main dependent variable of interest was the national incarceration rate. Various other measures (e.g., prison admissions by race, a measure of drug law enforcement, the percent of the population that was Black, homicide rate) were also examined.

Criminal threat narratives in the media rose between 1963 and 1988, coinciding with increases in the homicide rate. There was a close relationship between criminal threat narratives and change in the incarceration rate (r = 0.64), but no relationship over time between changes in the incarceration rate and the rate of political or economic threat narratives in the media. The homicide rate is also related to the long-term increases in the incarceration rate. Analyses suggest that “roughly half of the effect of homicide rates on incarceration can be attributed to criminal threat narratives” (p. 575).  

Conclusion: The findings suggest that to understand the manner in which mass imprisonment policies developed, one has to examine the manner in which the group who was disproportionately subjected to mass imprisonment in the US – Black Americans – was portrayed in media stories. Crime threat – but not political or economic threat – appears to be critical in understanding the growth of imprisonment in the final third of the 20^th century in the US.  Furthermore, the impact of serious crime – operationalized in terms of the homicide rate – on imprisonment was not only direct, but also was intimately intertwined with the view that Black Americans posed a threat to majority Americans in the area of crime.

Reference: Duxbury, Scott W. (2023). A Threatening Tone: Homicide, Racial Threat Narratives, and the Historical Growth of Incarceration in the United States, 1926-2016. Social Forces, 102, 561-585.

Item 3

Generalizations about whether it is good policy to have police serving the role of “school resource officers” cannot usefully be made. There are too many dimensions on which their deployment varies and too many different goals and groups who are affected by them to come to simple conclusions. 

“The presence of police officers in schools has become commonplace in recent decades” (p. 318), but what these school resource officers (SROs) do in the schools varies considerably. This paper examines the available research on school resource officers and notes that “much of what is documented in this area comes from small scale, localized studies of persons associated with schools that contain SROs” (p. 334).

“The role that an SRO plays largely depends on the officer, the school district, and their shared goals” (p. 320). However, in this review of 31 studies of perceptions and attitudes related to SROs, most SROs themselves defined their primary duty as being law enforcement and ensuring the safety of students. It is important to note that students in these studies often believed that SROs increased school safety and helped them feel safe. For example, in a study of SROs in schools in Edmonton, Alberta (Canada), not included in this review, it was found that “Regardless of race, sexual orientation and disability status, most students reported that their SRO made them feel safe at school.” Teachers, as well, saw SROs as contributing to school safety. At the same time, in some schools, students did express concern about the manner in which SROs enforced rules.

One feature of SRO programs mentioned in a number of studies covered by this review was that people perceived there to be an advantage of having an SRO physically present in the school and available if there were problems. Hence it was suggested that “it may be necessary for officers to be assigned to a specific school (or a small number of schools) where they can maintain a more consistent relationship with the student and faculty bodies” (p. 333). It may be that having a relationship with specific SROs reduces the likelihood both of discriminatory behaviour on the part of the SROs and of the SRO being seen as discriminating against certain groups. In the Edmonton study, for example, which included “interviews and surveys with over 11,000 students, 4,000 parents and 650 teachers”, it was found that “Regardless of race, sexual orientation and self-reported disability status, students and parents were much more likely to report positive experiences with their SRO (approximately 45 per cent of all respondents) than negative experiences (approximately 7 per cent of all respondents). Positive experiences included feelings of safety, assistance with victimization incidents, assistance with personal problems, informal conflict resolution, mentorship, legal education, and innovative strategies for discipline and reform.” That said, however, it is important to put the positive experiences in context. The Edmonton study also found that “Black and Indigenous students were somewhat more likely to support suspending the program than respondents from other racial backgrounds [and] were also more likely to report negative experiences with SROs, including allegations of oversurveillance, targeting and unfair disciplinary decisions.”

These papers did not examine whether there were changes in the frequency of incidents of serious violence that could be attributed to the SRO program. Nor did they look at whether the SRO program changed the likelihood of youths being charged with offences that would lead to suspensions and/or criminal charges.

Conclusion: In the end, these two studies demonstrated that if SROs programs are implemented in a community “We need to develop an evidence-based understanding of the best ways in which to incorporate these personnel in a way that does not harm students” (p. 334). Almost certainly, this means that the actual implementation in the local setting should be evaluated as an integral part of the implementation plan to ensure that the program is serving the functions it was designed to serve with minimal harmful side effects. Evidence-free assertions on any side of this debate are not helpful.

References: Almanza, Matthew, Makayla Mason and Chris Melde (2023). Perceptions of School Resource Officers. Criminal Justice Review 48(3), 318-338. Samuels-Wortley, Kanika et al. “Police in schools has long been a topic of debate. In Alberta, at least, the students have spoken.” Globe & Mail, 1 September 2003.

Item 4

Being suspended from school increases the likelihood that a youth will be arrested. Black and Hispanic students are more likely to be suspended than are White students, and one of the most important harms from this decision – subsequent arrest – is most likely to accrue to Black and Hispanic students.

There is substantial evidence that being suspended from school increases the likelihood that a student will subsequently commit offences and/or be arrested (Criminological Highlights 15(6)#1, 16(6)#1, 18(6)#3). This paper demonstrates that the impact of suspension on arrest is more serious for Black and Hispanic students than it is for White students.

“One of the most pervasive stereotypes in American culture is the stereotype linking Blackness to criminality” (p. 626). There are also data suggesting that teachers are more likely to suspect that Black students will misbehave.  Similar data exist with respect to Hispanic students. Given that the association between race/ethnicity and crime, this study examines whether the harmful impact of school suspension – and implicit labelling of the suspended student as a “criminal” – is likely to lead to more serious consequences for Black and Hispanic students than it is for White students.

The study uses data from a large longitudinal survey of American students born in the early 1980s. It examined data from 6 waves of data collected between 1997 and 2002 when, typically, the students were at the age when they were finishing or had finished high school. The main dependent variable was whether the youth had been arrested as a function of being suspended from school. Various factors were controlled statistically (e.g., whether the youth was from an inner city, self-report criminal offending, alcohol and drug use, socioeconomic background, etc.).

Not surprisingly, students who had been suspended had a considerably increased likelihood of being arrested compared to those who had not been suspended. However, “The impact of suspension has different consequences for arrest across racial/ethnic groups; Black and Hispanic students are more likely to experience an arrest in the following year compared to White students” (p. 638). For both Black and Hispanic students, being suspended from school increased substantially the probability of arrest. For White students, there was no such effect on the probability of arrest and in some tests, suspension may have decreased the likelihood of arrest (p. 640). 

Conclusion: It is possible that “Because suspension can be understood as congruent with racist stereotypes about Black and Hispanic students but stereotype incongruent with those pertaining to White students, the labelling process and secondary sanctioning may operate [in a different manner] for students by student race/ethnicity” (p. 646). One of the most pervasive social problems in the U.S. for the past several decades has been the racial disproportionality in criminal legal system involvement. Although many drivers for this disproportionality exist, this study’s findings suggest that the use of school suspension is likely to be one driver” (p. 648). Given that suspension does not make schools safer, it is hard to think of a justification for the practice, given its negative impact on students who are likely already disadvantaged.

Reference: Fisher, Benjamin W. and Alex O. Widdowson (2023) Racial and Ethnic Differences in the Consequences of School Suspension for Arrest. Criminology, 61, 662-653.

Item 5

Politics and policing are linked: White American police officers who favour the Republican party are more likely than White officers who favour the Democrats to search Black motorists whom they have stopped for an apparent driving infraction.

Policing and politics are intertwined. It is not unheard of for police officers to enter politics in the US or in Canada. And, in many jurisdictions, views about how the criminal justice system should operate vary across political parties. Given that police officers exercise an enormous amount of discretion in their interactions with citizens, it would not be surprising to find that their decisions on whom to investigate are correlated with their political orientation.

“Racial disparities in policing are well documented and extraordinarily persistent” (p. 637). In traffic stops there are often two separate decisions: whether to pull a motorist over for an apparent violation and whether to search the vehicle and/or driver following that stop. Racial disparities have been found at both of these decision points.

This paper examines the association between partisan identity – whether the police officer identifies as Republican or Democrat – and the officer’s decisions in the context of a traffic stop. It uses data on over 5 million traffic stops between 2012 and 2020 by the Florida Highway Patrol. The police data included, among other things, three key variables: whether the stop led to a search and the race of the person stopped and the race of the officer. The party affiliation of the police officer involved in each stop (i.e., whether the officer was a registered Republican or Democrat or was Independent/other) was obtained from Florida voter registration data. Incidents were examined that involved Black, Hispanic, and White officers and motorists. Police officers were more likely to identify as Republicans than did most residents of Florida.

The decision to stop a motorist was unrelated to the party affiliation of the police officer.  Overall, searches of motorists occurred relatively rarely. However, Black motorists were more likely to be searched than were White motorists. In addition, the disparity in the likelihood of a Black vs. a White motorist being searched was considerably larger for Republican police officers, whereas police officers who identify as Democratic were equally likely to search Hispanic and White motorists whom they had stopped. Republican police officers, however, were more likely to search Hispanic motorists whom they had stopped. Black and Hispanic police officers, on the other hand, appeared to decide on whether to search a motorist in a manner that was unrelated to their party affiliation.

When other control factors were included (e.g., the time of day of the stop, the county, the purpose of the stop), the results were consistent. “White Republican officers exhibit a larger Black/White racial disparity than White Democrats across racial and economic contexts and in counties with high and low crime rates” (p. 670).  However, “After Trump’s election in 2016, White officers exhibited a substantial increase in their likelihood of searching Black motorists relative to White motorists” (p. 672).  More detailed analyses demonstrated that this effect was due to changes in the likelihood that White officers would search Black motorists; and it was the result of a different propensity to search Black motorists among police officers who were hired after 2016.

Conclusion: The findings demonstrate that political party affiliation is related to front line decisions by police officers that affect Black and White motorists. Typically, we think of party politics and criminal justice in terms of how competing legislative agendas change criminal justice institutions and how these changes affect racial disparities. This study demonstrates that party politics affects behaviour directly at the street level.

Reference: Donahue, Samuel Thomas (2023). The Politics of Police. American Sociological Review, 88(4), 656-680.

Item 6

People with criminal records experience special obstacles when looking for rental housing. They are screened out of the rental market even before they have a chance to describe what their offences were and when they were convicted.

Previous research has linked eviction from one’s home and homelessness with offending (e.g., Criminological Highlights 20(4)#4, 1(2)#6). This paper examines the relationship between having a criminal record and the ability to find housing. The link is important for at least two separate reasons: people who already have been punished for their crimes are, in effect, receiving an additional punishment. But in addition, lack of stable housing is likely to be linked to reoffending.

In many cities, there is a shortage of affordable rental housing. Many large property management companies routinely do background checks on applicants for rental housing. Hence it is likely that landlords often have access to information about an applicant’s involvement with the law. Given that in the US it is estimated that 8.1% of the adult population has a felony record, the potential impact of a record on acquiring affordable housing could be substantial.

In 2018-2019, each of 924 landlords who had advertised a vacant apartment, received, as part of this study, two emails from people who expressed interest in renting the apartment. In one of the emails, the writer indicated that he had a criminal record and asked whether he would still qualify to rent the apartment. The other email was identical except that it did not mention criminal record. Race (Black, Hispanic, White) and gender of the rental applicant were manipulated through variation in the applicant’s name in the email. The main dependent variable was whether the landlord responded positively to the inquiry (as opposed to saying that the apartment was no longer available or not responding at all to the email).

For all 6 combinations of race and gender, those with no criminal record were more likely to receive a positive response from the landlord than were those without a criminal record (61% positive responses vs. 49%, pooled across race and gender). But in addition, “the effect of the criminal record does not vary significantly by the race or gender of the applicant” (p. 717). Independent of these findings, however, Black and Hispanic men were significantly less likely to receive a positive response from the landlord than were White men. 

Neighbourhoods were classified as “gentrifying” if, in recent years, they showed increases in both the proportion of adult residents with a bachelor’s degree and in property values. The findings demonstrate that the effect of criminal record on the likelihood of receiving a positive response from the landlord was greater in the gentrifying neighbourhoods than in the non-gentrifying neighbourhoods.

Other data suggest that “criminal record matters most [in reducing the likelihood that an applicant will receive a positive response from a landlord] where levels of disadvantage in the neighbourhood are declining and opportunities for economic prosperity are increasing for housing providers” (p. 719). Furthermore, the [effect of the criminal record in blocking access to housing] “is larger in neighbourhoods with a shrinking Black population and smaller in neighbourhoods with a growing Black population” (p. 720).

Conclusion: This paper shows that a criminal record makes it difficult for people to get rental housing, just as previous research has shown that having a criminal record reduces the likelihood of being considered for employment. These findings demonstrate that crime-free re-integration into society is, in part, directly under the control of the community that creates these barriers. Lack of housing and lack of employment are each related to reoffending. It is not easy to overcome the stigma of a criminal record. But clearly it is in society’s interest to explore ways in which these criminogenic barriers can be eliminated.

Reference: DeMarco, Laura M. (2023). Criminal record stigma, race, and neighbourhood inequality. Criminology, 61, 705-730.

Item 7

Employment policies that prohibit the use of criminal records may not accomplish what they are designed to do.  In fact, they may make it more difficult for certain groups of people (e.g., black men) to get jobs even if they do not have criminal records.

Because so many Americans – especially those from racial minorities – have criminal records, there have been numerous attempts, at the state level, to regulate the use of criminal records in deciding on whether someone should be hired. The state laws that govern the legal use of criminal records vary considerably in terms of their scope: the offences that are covered, whether they address convictions only and not arrests, and the employment positions that are regulated. 

Concern has been expressed, however, that these policies “can inadvertently exacerbate racial disparities by decreasing the likelihood that employers will hire young men from racial minorities” (p. 1124). The concern is that employers may “discriminate against groups deemed likely to have records – namely black men – as a proxy for criminal history” (p. 1126). This paper examines the association between state level policies on the use of criminal records by employers and the employment of men with and without criminal records.

Data from a large longitudinal study of disproportionately disadvantaged families with children born in 20 US cities between 1998 and 2000 were used in the study. As part of the original study, data on fathers’ involvement in the criminal justice system were collected as well as their involvement in paid work. State policies related to barriers to employment for individuals with criminal records were assessed and an index was created to describe the degree to which criminal record information could be used in employment decisions. States varied considerably in how protective they were, ranging from states in which there were no restrictions on the use of criminal records to states that had restrictions on the use of records in 6 different areas. At the time they were interviewed, these fathers lived in 46 different states.

The overall results demonstrated that living in a state that had policies in place to protect people from discrimination because of their criminal record was negatively associated with employment for black fathers both with and without criminal records. For white fathers, the polices made little difference. For black fathers, however, having a record and living in one of the states with policies designed to protect job applicants from having their criminal history used to their disadvantage was associated with a considerably lower likelihood of being employed. These findings reflect broader racial disparities. All black fathers were less likely to report having a job if they lived in a state with more employment discrimination protections in place, including those with no history of criminal justice contact. In fact, black fathers living in states without regulations on the use of criminal records were just about as likely to be employed as white fathers. In states where records were supposed to be protected, however, both black fathers with criminal records and their counterparts with no criminal justice history reported substantially lower levels of employment.

Conclusion: “These findings… contribute to a growing body of research warning that policies regulating the use of records without addressing underlying issues of racism in the criminal justice system and labour market may be ineffectual or even exacerbate existing disparities…. Putting employment policies on the books may be insufficient to overcome existing patterns of discrimination” (p. 1139).

Reference: Emory, Allison Dwyer (2023). Protective State Policies and the Employment of Fathers with Criminal Records. Social Problems, 70, 1123-1143.

Item 8

Attempts to control the use of imprisonment through policy changes targeting prosecutorial discretion are inherently unstable: Prosecutors often have at their discretion “uniquely powerful charging and plea-bargaining tools” (p. 1312) that can be used to target politically vulnerable groups.

The traditional view of the prosecutor is that of a “law enforcer, advocating on behalf of the state for conviction” (p. 1313). More recently it is recognized that prosecutors “wield significant power via charging and plea-bargaining tactics to lock in sanction outcomes prior to formal sentencing” (p. 1313). This is very clear in the case of the decisions to prosecute cases with mandatory minimum penalties.

In Canada, the availability of prosecutorial power is also evident in the decision on whether to proceed by way of summary conviction or indictment in cases in which the law provides this choice to the prosecutor. Typically not only is the maximum available prison sentence considerably longer in cases proceeded on by way of indictment (compared to by way of summary conviction) but so is the length, if any, of a mandatory minimum sentence.

This study looks at the way in which US federal prosecutorial discretion in two district courts changed between the early 2010s and the early Trump era (2017-2019), focusing largely on immigration and drug trafficking cases. Prosecutors and defence counsel were interviewed. In addition, data on the processing of federal cases were examined.

Neither of these districts was on the southwestern US border (where immigration laws may be especially important) and there were no important changes in the formal law during the study period. Typically, in the first part of the 2010s, immigrant cases where the accused did not have a serious record ended up with simple deportation. In the Trump era, however, the portion of those with little or no criminal record who were charged with immigration offences increased. In one district, the proportion of those without serious criminal records increased from 13% in 2015 to 33% in 2018. In the second district, similar increases were found. Illegal immigrants who used false identities to get or maintain jobs were much more likely, in the Trump era, to be charged with offences carrying mandatory minimum sentences. Similarly, charging practices changed dramatically with drug offences such that the proportion of drug defendants facing mandatory minimum penalties increased substantially.

During the Trump era, “frontline prosecutors used the statutory tools at their disposal to implement the regressive war-on-crime [procedures that were favoured by Trump’s Attorneys General]” (p. 1328). Under the previous (Obama) administration, prosecutors, in these same locations, had reduced the use of these punitive procedures.

Conclusion: It is often the case that groups that favour or oppose the use of imprisonment for many offences (e.g., drug and immigration offences) will focus on what the formal law appears to require. This study “highlights the key role frontline prosecutors play in maintaining mass incarceration” [in the U.S.] (p. 1329). Each of the two districts that was studied became “more restrained in the how they prosecuted drug offences [during the Obama period but] that restraint quickly disappeared with the change of administration” (p. 1329). Long term change, then, is not likely to be easy to accomplish if the orientation of prosecutors is a direct function of which political party is in power.

Reference: Lynch, Mona (2023) Prosecutors as Punishers: A Case Study of Trump-era Practices. Punishment & Society, 25(5), 1312-1333.

This issue of Criminological Highlights was prepared by Anthony Doob, Rosemary Gartner, Maria Jung, Tyler King, Audrey Macklin, Jane Sprott, and Danielle Van Wagner.