Criminological Highlights Vol. 20, No. 4 - January 2023

View this issue as a PDF: PDF iconCriminological Highlights Volume 20 No. 4 

This issue of Criminological Highlights addresses the following questions:

  1. What are the risks when financial obligations are imposed on people who have been found guilty?
  2. When does the supervision of people on probation become counter-productive?
  3. Are prisoners suffering mental illness targeted for placement in solitary confinement?
  4. How can the impact of victimization on feelings of being unsafe be reduced?
  5. How do housing policies contribute to crime?
  6. Does “neighbourhood policing” affect crime and how the police respond to it?
  7. What concerns need to be addressed when thinking about implementing a program that has been “proven effective” in some other community?
  8. What happens to refugees in a country after there is a terrorist attack?

Court orders that turn minor offenders into debtors (to the state) appear to be more likely to create negative than positive effects.  Court-created debt does not reliably create or deter new crime and, typically, little money is collected.  But court-created debt does end up re-criminalizing low-income debtors as a result of debt-collection attempts. 

Many court systems include monetary penalties in their arsenal of punishments.  In Canada’s Criminal Code, for example, there are fines (S. 734), victim payments (S. 732), and restitution orders (s. 738).  But in addition, people cannot apply for a “record suppression” (pardon) until a specified period of time has elapsed after all penalties (including the payment of fines, victim surcharges, etc.) have been discharged.  Financial penalties can, therefore, have long term consequences by delaying the ability of those who have offended to rid themselves of the label “offender.”

This paper examined the consequences and impact of financial penalties on people found guilty in misdemeanor court in a county in Oklahoma, USA.  In this county, the largest financial obligations imposed on people came from court costs, supervision fees on those found guilty, and prosecution fees.  Probation supervision fees, for example, averaged $511 per person.  People in misdemeanor court (both in- and out-of-custody), mostly (86%) defended by public defenders, who had been assessed fines and fees were randomly assigned to one of two conditions: normal treatment (control) and a treatment condition where the study paid for all of their financial obligations related to the case (totaling an average of US$1,779 per person).

There was some evidence that in the first three months after formal conviction, people whose fines and fees had been paid by the study were less likely to be charged with a new offence and to have new criminal convictions.  However, one year after conviction, there were no overall effects on these two measures or on new stays in jail. For those who were not employed at the time of their original court hearing (about half of the sample), jail bookings were considerably higher for those whose debt was not paid by the study.   The debt-relief treatment was dramatic when one looks at measures related to the original conviction.  Within a few months, those whose debt had not been eliminated were considerably more likely to have new warrants issued for their arrest and new debts added to the earlier ones (e.g., financial penalties for the non-payment of the previous debts).  In addition, their debts were more likely to have been turned over to private debt collectors and courts were considerably more likely to order that any tax refund owing to the person should be paid to the county.  But debt-relief did not have lasting effects on new criminal behaviour.

Those whose debt was not assumed by the study were not likely to pay down their debts to any significant degree. 80% made no payments whatsoever, and payments averaged about US$43, or less than 4% of the average fee levied on those in the study.

Conclusion: The results illustrate how “poor people become subject to enduring criminal justice control…. Fines and fees criminalize poverty by prolonging criminal court involvement for those unable to pay” (p. 545-6).  Court efforts to collect outstanding court-created debt were largely futile. But in addition, the data demonstrate that “mass waiver of fines and fees would have no effect on recidivism or court finances, but it would limit court supervision rooted in poverty” (p. 547).  It would, however, “eliminate a large but ineffective cost-recovery bureaucracy” (p. 547).  But in addition, the study demonstrates that attempting to get blood out of a stone is not an effective crime control strategy.

Reference: Pager, Devah, Rebecca Goldstein, Helen Ho, and Bruce Western (2022).  Criminalizing Poverty: The Consequences of Court Fees in a Randomized Experiment.  American Sociological Review, 87(3), 529-553.

Back to top

Incarceration may be the most punitive of penalties available in non-execution jurisdictions, but the constraints, control, and punitive aspects of probation, as an alternative, should be considered carefully when this sanction is imposed.  The use of probation for those convicted of offences can ‘wrap them up, smother them and incite them to conduct themselves in particular ways’ (p. 812).

On an average day in 2020, there were, in the Canadian province of Ontario, 6.2 times as many people serving probation sentences as there were serving sentences in federal or provincial prisons in the province.  Though probation, as an alternative to imprisonment, might have been seen as the “ideal penal form of the Progressive Era” (p. 800), there have been concerns related, in part, to the fact that the sanction includes conflicting goals.

In the US, the number of people on probation more than tripled between 1980 and 2016. Typically, those on probation have at least 20 mandatory conditions on them in addition to, often, various legal financial obligations.  Not surprisingly, there is a high revocation rate in that country, often resulting in imprisonment. Those serving sentences often prefer short jail sentences to probation (see Criminological Highlights 19(1)#6).

In this study of the views of 156 adults on probation and 172 probation officers, one of the more important cross-cutting themes was the variability in the approaches taken by probation officers and the manner in which probationers experienced their supervisory relationships. “In very tangible ways, probation strips adults of their independence and the ability to make autonomous choices about their lives (where to live and work, how to spend their time, with whom to associate)” (p. 806).  These experiences were sometimes viewed as infantilizing. Probation officers often saw themselves as resource brokers – finding programs for those being supervised. But simultaneously, many officers would frame programs as a form of punishment.  Often probation requirements created conflicts such as when probationers had to appear at specific times for drug testing and yet were expected to be working at the same time. The impossibility of meeting all the demands of supervision, then, made those being supervised very aware of the arbitrary manner in which ‘violations’ were enforced. 

However, for some people, “probation was perceived as a lifeline, providing caring and transformative resources that they had been unable to access before conviction” (p. 811).  Hence variation in treatment appeared to be the norm, rather than the exception.

Conclusion:  Part of the problem with probation appears to be that there is little thought about the appropriate relationship between the person being supervised and the supervisor, what programs and services should be provided or required and how these requirements should be balanced with the need to attend work and earn money. But overall, it may be that without clear priorities, goals and methods of achieving those goals (see Criminological Highlights 9(6)#3, 19(4)#2), it is inevitable that probation appears to be a criminal justice sanction with many contradictions that interfere with the potentially positive aspects of it identified by both probationers and officers.  As suggested by a well-known lay criminologist, “If you don’t know where you are going, you may not get there.”

Reference: Phelps, Michelle S. and Ebony L. Ruhland (2022). Governing Marginality: Coercion and Care in Probation.  Social Problems, 69, 799-816.

Back to top

Prisoners experiencing mental illness are more likely than other prisoners to be subjected to periods of solitary confinement while in prison.  This increased risk of being placed in solitary confinement is not fully explained by prison misconduct on the part of the prisoner.

Concerns have been raised about whether solitary confinement is being used in prisons “for ‘nuisance’ [prisoners] rather than for those who pose a legitimate threat to prison safety and security” (p. 772) (See Criminological Highlights 20(3)#5).  Canadian findings, for example, demonstrate that prisoners who are frequently placed in its federal solitary confinement cells are dramatically more likely than those who are placed in solitary less often to have been identified as having mental health needs.

An alternative explanation for the association between mental illness and time in solitary confinement is, of course, that solitary confinement creates mental health problems (see Criminological Highlights 19(6)#4). Importantly, being in solitary confinement may be both the result of a prisoner experiencing mental health symptoms and also a cause of deteriorating mental health.

This study looks at whether incarcerated persons with mental illness are more likely to experience time in solitary confinement than similar prisoners without mental illness.  The challenge, of course, is to ensure that the comparison group – ‘similar’ prisoners without mental illness – really is similar.  Records of prisoners with and without mental health diagnoses were matched on about 40 factors measured at admission or soon thereafter. These included arrest records, experience with prison or community punishment, current offence, and various background characteristics.  Using these factors to calculate “propensity scores” (propensity to have mental health challenges), the two groups could be compared to see if actual mental health made a difference above and beyond these factors in the likelihood of being sent to solitary confinement.   

The results show that having a mental health challenge predicted both placements in solitary confinement and the total number of days in solitary confinement.

Three ‘mediators’ were also examined: whether the prisoner had been a victim of a violent or property crime, whether the prisoner had communicated to prison authorities that he was afraid for his life, and whether the prisoner had been disciplined for violence. These measures were obtained during the first 60 days.  It turns out that disciplinary (violent and non-violent) infractions by the prisoner reduced, slightly, the strength of the relationship between mental illness and placement in solitary confinement.  But the effect of mental illness on the likelihood of being placed in solitary confinement remained after controlling for disciplinary infractions.  Victimization concerns about the prisoner did not reduce the strength of the relationship.

Conclusion: Prisoners experiencing mental illness “are at increased risk for being sent to solitary confinement…. [This risk] is present for a variety of mental disorders and is only partially explained by prison misconduct” (p. 768).  Given that time in solitary confinement is likely to increase the likelihood of a prisoner experiencing mental illness, one can easily see that multiple stays in solitary confinement may be a natural outcome of prison authorities responding to the mental illness of a prisoner rather than the prisoner’s actual behaviour.

Reference: Siennick, Sonja E., Mayra Picon, Jennifer M. Brown, and Daniel P. Mears (2022), Revising and Unpacking the Mental Illness and Solitary Confinement Relationship.  Justice Quarterly, 39(4), 882-801.

Back to top

It is well established that in many countries victims of crime are likely to feel unsafe and vulnerable to additional victimization.  However, in countries with high levels of welfare provision, low levels of income inequality and high levels of trust in the criminal justice system, the strength of the relationship between victimization and feelings of unsafety is considerably reduced.

There is reason to believe that the effect of victimization on feelings of unsafety may depend on the social system of the community in which victimization occurs. This study examines how “the societal context, in terms of inequality, welfare policies, trust in the criminal justice system and welfare regimes moderates the impact of victimization on feelings of unsafety” (p. 1306).

The study uses data from three rounds of the European Social Survey, using data from 108,792 respondents in 21 European countries.  The focus was on understanding variation in responses to the question “How safe do you – or would you – feel walking alone in this area after dark?” (p. 1312). Victimization was estimated from the question, “Have you, or a member of your household, been the victim of a burglary or assault in the last 5 years?” (p. 1313).   The 21 European countries were grouped according to the degree to which their welfare policies were designed to protect people from social risks (e.g., unemployment), the level of inequality that exists in the country, and the overall trust that citizens have in the criminal justice system.  The Nordic countries were at the top of this scale; the Eastern European countries were at the bottom.

A large number of variables were controlled statistically.  These included age, gender, education, various income measures, health status and the size of the town or city that a person lived in.  Respondents were also asked how much they trust the police and legal system.

Across countries, as trust in the criminal justice system increased, the level of feelings of not being safe decreased.  However, the decrease in feelings of not being safe as trust increased was more dramatic for those who experienced victimization.  Said differently, having been victimized had a larger impact on feelings of unsafety for those who had low levels of trust in the criminal justice system.  Those who lived in countries with high levels of economic inequality were more likely to report that they felt unsafe.  The difference between victims and non-victims was larger for those living in countries with high levels of economic inequality than it was for those who lived in countries with less inequality.  The findings were similar when looking across countries that varied in their level of expenditure for social protection.  Feelings of not being safe decreased as expenditures for social protection increased, and in countries with high levels of expenditures for social protection, the difference between feelings of unsafety for victims and non-victims was quite small.

Respondents in Eastern Europe showed the highest rate of feelings of unsafety for both victims and non-victims. Furthermore, the difference between the ratings of unsafety for victims vs. non-victims was highest in this region.  At the other extreme, victims and non-victims hardly differed in the Nordic countries and they showed the lowest level of feelings of unsafety.

Conclusion:  Overall, it seems that victimization has less impact on people if they live in countries with high levels of social protection and low levels of income inequality. This may be because they believe that they will be taken care of and will be able to cope if problems arise.  Clearly the societal context in which victimization takes place is important in understanding people’s feelings of being vulnerable to crime as well as the impact that being victimized has on their future feelings of unsafety. “Combating fear of crime is not primarily a matter of reducing the objective risk of being victimized, but rather a question of building good governance and better services for victims of crime” (p. 1321).

Reference: Ejrnæ, Anders and Rune H. Scherg (2022).  The Impact of Victimization on Feelings of Unsafety in Different Welfare Regimes. European Journal of Criminology, 19(6) 1304-1326.

Back to top

Many cities in many countries are experiencing a housing crisis.  A housing crisis can lead to an increase in crime if communities allow people to be evicted from the places where they are living.

Prior research has found that people who are threatened with eviction are more likely to be involved in criminal offending than those who are not. Not surprisingly, homeless youths are also more likely than other youths to be involved in crime (Criminological Highlights 1(2)#6).  This study goes one step further and attempts to determine whether eviction from one’s home contributes directly to crime.

The study examines the effect of eviction on three types of crime: Utilitarian crime (theft, fraud, robbery), other violent crime (homicide, assault, sexual offences, threats), and drug crimes. The data come from a database including all those people actually evicted from their housing in Sweden (excluding those, for example, who moved out on their own after being notified of a possible eviction) during the period 2009-2012 as well as a representative sample of Swedish adults who had not been evicted.  On the basis of information about gender, age, country of birth, education, household size, whether there had been a psychiatric diagnosis, income and receipt of social assistance, the probability of being evicted was calculated for each person. Then, those actually evicted were paired with people who had not been evicted but had the same “propensity” score (propensity to be evicted). 

Before matching, it was not surprising to find that those who were evicted had higher rates of all three types of crime than those in the general population throughout the period that was examined. 

The main findings examined conviction rates only for those with no previous criminal convictions.  Conviction rates for these two groups – those who experienced eviction and those who did not – were monitored for three years.  For all three categories of crime, those people who had been evicted from their previous homes had higher rates of crime during each of the three follow-up years.  The size of the effect appeared to be larger for “utilitarian” (property related) crimes than for violent or drug crimes.

The size of the effects was not large which may have been the result of examining only those people without a past history of criminal conviction.  Nevertheless, a number of additional analyses pointed to the likelihood that – for this group of people who had never been convicted of a criminal offence – being evicted from one’s home caused them to behave in a manner that led to criminal convictions.

Conclusion:  Previous research (Criminological Highlights 15(6)#5) has found that stable ordinary housing can reduce offending by those with mental health challenges.  This paper complements these findings by demonstrating that for people who have never been convicted of a criminal offence, being evicted from one’s home leads to increased utilitarian, violent, and drug crime that lasts at least three years after the eviction has taken place.

Reference: Alm, Susanne and Olof Bäckman (2022).  ‘When It Rains, It Pours’: Housing Evictions and Criminal Convictions in Sweden. European Journal of Criminology, 19(4), 612-631.

Back to top

“Neighbourhood policing” – a policing strategy that encourages informal interactions between the police and community members – reduced arrests in New York City. However, the strategy had little impact on crime or the racial disparities of the arrests that were made.

When a “neighbourhood policing” program was started in New York City in 2015, the then mayor said that its purpose was, in part, “to build relationships and deepen trust” (p. 471).  The idea was that for a certain portion of officers, their primary focus was to be on community engagement.  They walked their beat and didn’t, for at least part of their shift, have responsibility for emergency response.

This paper examines the impact of this program on four dependent variables: reported crime, civilian complaints of police misconduct, low level arrests, and racial disparities in low-level arrests.  The hope was that by “improving cooperation with community members, they would solve and deter more crime” (p. 472). 

The timing of the implementation of the program allowed fairly strong inferences to be made about its impact because the program was implemented slowly on a precinct-by-precinct basis between May 2015 and October 2018.  Using data on New York City’s 76 residential precincts between January 2006 and September 2019, the researchers were able to estimate for each of the precincts what would likely have happened to the four major dependent variables if the program had not been implemented.  This could be done because comparison precincts (without the program) were available during the period when the program was slowly being rolled out across New York City.  This approach allowed the researchers to start from the trend line of crime before implementation and then to see what happened in each of the precincts when the program was implemented, compared to an estimate of what would have been expected if there had been no program (based on the precincts that did not get the program at that time).

Neighbourhood policing had no effect on either violent or property crime. Civilian complaints, however, were significantly higher in the period 5-7 months after implementation. However, 8 months after implementation, civilian complaints were back to what would have been expected had the program not existed. Generally, the increase was due to complaints of “discourtesy/abuse of authority” (p. 485).

Proactive arrests (arrests initiated by the police officer on the beat rather than as a result of calls to the police) decreased when the program came into effect. In general, it appears that misdemeanor arrests, more generally, also declined somewhat after implementation.  In fact, “a sharp decrease in arrests occurred the month neighbourhood policing was introduced to each precinct, and since each precinct adopted the policy at different times, this break supports the conclusion… that it is neighbourhood policing and not a city-wide change that drove the arrest decline” (p.485).  However, there was no impact whatsoever on the difference between the Black and White proactive arrest rates.

Conclusion: Two of the more interesting findings of the study were that the implementation of neighbourhood policing reduced low level arrests and had no simultaneous impact on crime.  Said differently, the findings support the idea that there is little, if any, likelihood that crime will increase if a policy comes into place that encourages police to find solutions for minor community problems without using their powers of arrest.   In this context, it is important to note the evidence suggested that “neighbourhood policing’s arrest reduction was greatest in poorer neighbourhoods, areas traditionally exposed to the highest levels of discretionary arrests” (p. 491). At the same time, it is important to keep in mind that this policy did not reduce the racial disparity of arrests.

Reference: Beck, Brenden, Joseph Antonelli, and Gabriela Piñeros (2022). Effects of New York City’s Neighbourhood Policing Policy. Police Quarterly, 25(4), 470-496.

Back to top

A program for probationers that is effective in one location may not be effective in another.  In this study, a program designed to address offenders’ criminogenic needs reduced reoffending only in a location that, previously, had fewer resources for such offenders and limited history of cross-system collaboration.

Just because a program has been shown to be effective ‘somewhere’ does not mean it will be effective everywhere.  The implementation of a program may vary across location, and, in addition, what is ‘normal treatment’ (‘business as usual’) can vary across locations. 

In this (Oregon) study, people who had substance abuse problems, had apparently committed a property offence, and would normally have gone to prison were selected for special treatment if the county’s prosecutor was content that incarceration wasn’t necessary.  They were then randomly assigned either to probation as usual (the ‘control’) or a special treatment program. In the control condition, programs were available, but not mandatory, and supervisors were not likely to have been trained to guide those being supervised to effective programs. In the treatment program, treatment began within a week, and was based on risk, etc., criteria (see Criminological Highlights, 9(6)#3&4; 19(4)#2). Furthermore, problems (e.g., missed appointments) were carefully monitored to ensure programs were completed.  The control participants had, in theory, access to the same programs, but there was less monitoring and mentoring. The main outcome measures related to arrests, convictions, and incarcerations.

In general, in County A, there were no differences between the treatment program and the control on the criminal justice outcome measures.  In County B, however, there were differences between the treatment and control conditions on arrest, conviction and incarceration for a new crime.

In understanding why the program apparently ‘worked’ in County B, but not County A, it should be noted the participants in the two counties looked very similar; hence it is unlikely that participant characteristics were responsible for this difference.  Much more likely is the fact that County A’s community corrections program was – prior to the study – much better resourced. Hence the difference between ‘treatment’ and ‘control’ was, perhaps, less dramatic in that location. In addition, prior to the study, there already was, in County A, a “long and well-established history of …  cross-system [prosecutor and corrections] collaboration” (p. 13). The beginning of this study created, in County B, a close collaboration between the prosecutor and the corrections people who were responsible for the program.  But there was no such close working relationship between prosecutors and corrections people with respect to the control group.

Conclusion. There are three straightforward lessons to be learned from this study.  First, program design and delivery in community corrections can be modified in a manner that effectively reduces reoffending.  In this case, a coordinated program of delivery of treatment programs to complex property offenders was effective in one location.  Second, formal changes to a program will not necessarily always reduce re-offending.  In this case, the ‘new’ program effectively changed reoffending in one location, but not another.  Third, it is important to evaluate new programs in the setting in which they are being implemented.  In this case, if the program had been evaluated only in County B, nobody would know that the setting was an important determinant of the program’s overall effectiveness.

Reference: McCart, Michael R., Jason E. Chapman, Zoe Alley, & Ashli J. Sheidow (2022). Randomized trial of a diversion program for property offenders with drug use. Journal of Criminal Justice, 79, 1-15.

Back to top

In a 10-day period in July 2016, three Islamist terrorist attacks occurred in France and Germany. Data from surveys that were in progress in Germany at the time of the attacks show that anti-refugee sentiment increased, and refugees experienced more discrimination and suffered declines in their mental health.

There is substantial research demonstrating that terrorist attacks harden the majority group’s attitudes against immigrant and refugee groups. This paper extends these findings by examining attitudes among the majority as well as among refugees and asylum seekers.

On 14 July 2016 in Nice, France, 86 people were killed in one terrorist incident. Within 10 days two serious incidents happened in Germany injuring 20 other people.  Before these incidents occurred, a survey (total n=959) of German citizens’ “negative emotions towards, risk assessments of, and perceived social distance to asylum seekers” (p. 953) was being carried out. For refugee respondents in Germany at that time, a survey (n=1038) was already collecting data on various aspects of their living experiences, mental health, and well-being.  In each survey, a substantial portion of the respondents (54% for the German citizens; 37% for the refugees) were surveyed before the events of July 2016.

After the terrorist events, German citizens showed a substantial increase in fear of and anger toward refugees.  They also reported that they believed there were greater risks associated with the refugees after the events. They were especially concerned about the threats they believed refugees pose to safety and social cohesion. Concerns about safety were expressed by 68% of the respondents before the events; the attacks increased that by 7%.   German citizens in this survey were also asked to rate their emotions towards other minority groups in Germany (Polish, Jewish, Italian and Turkish residents).  Feelings towards these groups were unaffected by the July 2016 events.

Refugees, after the July 2016 events, were more likely to report that they had experienced discrimination and felt less welcome than did comparable refugees interviewed before the July attacks.  “The surge in anti-refugee sentiment among the majority population matches refugees’ own experiences of being exposed to more discrimination and hostility in the aftermath of the attacks” (p. 961).  Interestingly, however, the refugees themselves did not indicate that they were more likely to worry about anti-immigrant attitudes following the July incidents. Only about 30% of respondents reported being somewhat or very concerned about these matters.  Their worries apparently focused more on their immediate problems: their dire economic situation and their future prospects in Germany and in their countries of origin. 

The impact of the terrorist events on refugees’ and asylum seekers’ mental health (measured with two separate indices) was immediate and dramatic.  Their mental health suffered most in the first five days after the attacks but tended to return to the pre-attack level after about 3-4 weeks. Not surprisingly, this trend over time corresponded with refugees’ reports of exposure to hostility and the period when they were least likely to feel welcome.

Conclusion: Exposure to terrorism from people perceived to be outsiders exacerbated anti-refugee sentiment, though it did not affect respondents’ views of other immigrant groups.  More important, perhaps, are the effects on the immigrant groups blamed for the terrorism.  The increase in their feelings of discrimination and hostility as well as their decreased mental health could well have affected other aspects of their lives with possible long-lasting impacts.

Reference:  Frey, Arun (2022). Getting under the Skin: The Impact of Terrorist Attacks on Native and Immigrant Sentiment.  Social Forces, 101(2), 943-973.

Back to top

Publication Type

Volume Number


Issue Number