Race, police stops, and perceptions of anti-Black police discrimination in Toronto, Canada over a quarter century

Abstract

Purpose

Black Canadians have a historically tenuous relationship with the police. Negative perceptions of the police held by Black people have traditionally resulted from high levels of police contact and perceived negative treatment during these encounters. Well-publicized instances of police violence involving Black civilians have also fostered hostility and mistrust of the police, often resulting in social unrest. Recently, in the wake of George Floyd's death at the hands of American police, people across Canada rallied in support of the Black Lives Matter social movement and calls to defund the police entered mainstream political consciousness. At the same time, police leaders have vehemently argued that racial bias within Canadian policing has been greatly reduced as the result of various reform efforts.

Design/methodology/approach

This paper explores the police racism debate in Canada through an analysis of three waves of survey data collected between 1994 and 2019.

Findings

Despite well-publicized reform efforts, the authors' findings demonstrate that little has changed over the past 25 years. Black people still report much higher rates of police stop and search activity than people from other racial backgrounds. Furthermore, racial disparities in negative police contact remain strongly significant after controlling for other theoretically relevant factors, including self-reported deviance and community crime levels. Finally, reflecting their negative experiences, most Black people still perceive Canadian law enforcement as racially biased. Nonetheless, the data do reveal one significant change: the proportion of white people who perceive police discrimination against Black people has increased dramatically over this same time period. The paper concludes by discussing the prospects of meaningful reform in light of the current findings.

Originality/value

This paper contributes to the literature on race and policing through an examination of 25 years of survey data across three waves of collection.

Citation

Wortley, S. and Owusu-Bempah, A. (2022), "Race, police stops, and perceptions of anti-Black police discrimination in Toronto, Canada over a quarter century", Policing: An International Journal, Vol. 45 No. 4, pp. 570-585. https://doi.org/10.1108/PIJPSM-11-2021-0157

Introduction

Despite an international reputation as a racially tolerant, multicultural society, Canada's Black population has long expressed concerns about discriminatory police treatment ranging from biased stop and search activities to police use of force (Maynard, 2017). As has been the case in the United States and the United Kingdom, the unjust police treatment of Black people has often been a catalyst for social unrest (Cobbina et al., 2016; Owusu-Bempah, 2014). Much of the controversy over race and policing has centered on Toronto, the county's largest and most racially diverse city. Historical analysis has illuminated how the police disproportionately applied public order offenses as a means of controlling Toronto's Black population in the early 20th century (Mosher, 1998). As Canada's immigration reforms of the 1960s opened the doors to immigrants from non-European countries, an influx of Black people arrived in Canada, many settling in the Greater Toronto Area (Milan and Tran, 2004; Chui et al., 2008). Shortly thereafter, tensions between Toronto's growing Black community and the police escalated into public protest following a series of police shootings involving Black men in the late 1970s (Owusu-Bempah, 2014).

The first of these high-profile incidents took place on August 9th, 1978, when a white police officer shot and killed Buddy Evans, a 24-year-old Black man, in a Toronto nightclub (Stasiulis, 1989). The officer's acquittal, following an 11-week inquest, prompted a rally organized by the Action Committee Against Racism. A year later, another Black man, 35-year-old Albert Johnson, was shot to death in his own home by two white officers (Stasiulis, 1989). This shooting again sparked protests. Dudley Laws, who would become an important figure in the fight against police discrimination in Canada, formed the Albert Johnson Defense Committee Against Police Brutality. In October of 1979, members and supporters of this Committee gathered outside city hall to protest the death of Johnson (Owusu-Bempah, 2014). The two officers involved in the Johnson shooting were ultimately acquitted of manslaughter charges. However, the Ontario Government responded to the demonstrations with proposed reforms – including the establishment of the Office of the Public Complaints Commissioner. This initiative, intended to provide an improved civilian complaint system, was viewed as biased in favor of the police and ultimately subject to great criticism from Toronto's Black communities (Owusu-Bempah, 2014).

A second series of police shootings involving Black men in the late 1980s prompted further community mobilization and unrest. Days after the 1988 police shooting of 44-year-old Lester Donaldson in his Toronto rooming house apartment, the Black Action Defense Committee (BADC) organized a demonstration in front of the police division where the suspect officer worked. On December 8, 1988, a Black teenager, 17-year-old Michael Wade Lawson, was shot in the back of the head by a Toronto-area police officer using a hollow-point bullet which had been banned under Ontario's Police Services Act (Jackson, 1994). Public demonstrations erupted after the acquittals of the officers involved. In response to this social unrest, the provincial government formed the Task Force on Race Relations and Policing, which was mandated to “address promptly the very serious concerns of visible minority communities respecting the interaction of the police community with their own” (Lewis, 1989). The Task Force documented perceptions of unjust treatment among racialized people in the province and made a series of recommendations for reform. These recommendations included the creation of a civilian oversight body, the Special Investigations Unit (SIU), tasked with investigating serious injuries and deaths that resulted from police activity (Lewis, 1989; Ontario, 2003; Wortley, 2006).

Despite the establishment of the SIU, police shootings of Black people in Toronto did not abate. On May 2nd 1992, as riots raged in Los Angeles following the acquittal of the officers involved in the beating of Rodney King, two Toronto police officers shot and killed a young Black man named Raymond Lawrence. Two days later, on May 4th, the Black Action Defense Committee organized a demonstration outside of the US Consulate in Toronto, drawing a crowd of 500 people (Croucher, 1997). What started as a peaceful demonstration turned destructive as other parties – including white skinheads – joined the protest (Vyhnak, 1992). Once again, social unrest prompted political action. Ontario Premier Bob Rae admitted that systemic racism was an emerging problem that warranted further investigation. A subsequent report concluded that racialized people, particularly Black Ontarians, experienced discrimination in policing and the administration of justice (Ontario Human Rights Commission, 2003). Later in 1992, in response to growing public concern, the Government established the Commission on Systemic Racism in the Ontario Criminal Justice System. The Commission's mandate was to undertake a comprehensive review of racism within the province's criminal justice system. In December 1995, the Commission delivered a 450-page report, detailing the scope and nature of police discrimination in the province and making dozens of recommendations for reform (Commission on Systemic Racism in the Ontario Criminal Justice System, 1995).

Throughout the 1990s and into the 2000s, public concern about police bias shifted to the issue of racial profiling and allegations that Toronto's Black community was disproportionately subject to unfair police stop, question and search activities. In 2002, the Toronto Star published its first investigation into race and policing in Ontario (Rankin et al., 2002a,b). The lead article entitled Police Target Black Drivers, showed that Black Torontonians accounted for almost 34% of all civilians charged with various traffic violations, even though they accounted for just 8% of Toronto's population. Police Chief Julian Fantino questioned the Star's analysis and findings, proclaiming “[w]e do not do profiling at this service” (Rankin et al., 2002a). Many Black Torontonians disagreed. The Star's work was pivotal in terms of propelling issues of race and police contact into the public consciousness. Once the denials had subsided, police officials began to promote various reforms. These reforms were centered around an initiative dubbed Project Charter, that brought together the Toronto Police Service (TPS), the Toronto Police Service Board, and the Ontario Human Rights Commission, to address issues related to diversity recruitment, anti-bias training, public education, community policing, and enhanced civilian oversight (Ontario Human Rights Commission, 2007; Culkier, 2014).

While the Toronto Police undertook what was considered by some to be performative work [1] in addressing racial discrimination, the Toronto Star continued its investigative journalism into race and policing (Owusu-Bempah, 2014). In 2010, the Star released another series entitled Race Matters, which documented their analysis of over 1.7 million “contact cards”, filled out by the Toronto Police Service between 2003 and 2008 (Rankin, 2010a, b). These data captured police encounters in which an officer documented the personal information of civilians for intelligence purposes [2]. The Star's analysis revealed that Black people were grossly over-represented in TPS “carding’ practices. Despite admitting that racial profiling existed when he took over as Toronto's police chief, Chief Bill Blair denied that the racial disparities observed in the Star's analysis could have resulted from police racial bias (Toronto Star, 2010).

As before, government action followed the public outcry over the racial disparities found in the Toronto police database. The Honorable Michael H. Tulloch, a judge on Ontario's Court of Appeal, was appointed to conduct a review into the practice of “carding”, otherwise known as “street checks”. This review led to the implementation of legislation designed to curb the practice (Tulloch, 2018). In the meantime, Toronto's Black communities continued to assert themselves in the streets. Black Lives Matter Toronto, the first BLM chapter outside of the United States, formed in 2014, drawing renewed attention to police killings and other forms of racial injustice. Subsequent police shootings of Black men, including Jermaine Carby and Andrew Loku, resulted in protests led by BLM leaders (Clarke, 2015; Reddekopp, 2017). These efforts were redoubled following the death of Regis Korchinski-Paquet, a Black woman, in the presence of Toronto Police officers. This incident took place on May 27, 2020, two days after George Floyd was killed in Minneapolis (Gillis, 2020). Once again, police and politicians responded, with Chief Mark Saunders, Toronto's first Black police chief, dropping to a knee with protestors and promising additional police reforms (Nickle, 2020).

Despite repeated cycles of injustice, social unrest, and government action, little research has examined whether the various initiatives aimed at reducing police discrimination have been impactful. Indeed, there has been no assessment of the implementation of the numerous recommendations stemming from more than a half dozen inquiries and taskforces, nor has there been a thorough evaluation of the extent to which specific initiatives, such as Project Charter, have reduced racial bias in policing and improved police-race relations (Culkier, 2014). This research aims to fill this gap by examining changes in public perceptions and experiences with the police in Toronto over the 25-year period between 1994 and 2019.

Research questions

Mirroring the situation in most American cities, the issue of anti-Black racism within policing has been a major social issue in Toronto, Canada for at least the past 5 decades. Over the past thirty years, several government commissions have explored the issue and tabled numerous recommendations designed to both reduce bias and improve the relationship between the police and Toronto's Black communities (Commission on Systemic Racism in the Ontario Criminal Justice System, 1995; Ontario Human Rights Commission, 2003, 2018) Popular anti-bias initiatives, implemented by the Toronto police, include anti-bias training, community policing, diversity hiring, and increased civilian oversight (Toronto Police Service, 2013; Culkier, 2014). Have these initiatives had an impact? Have these initiatives improved how Black civilians view the police and reduced racial disparities with respect to police outcomes? The analysis below addresses the following three research questions:

1. Have perceptions of anti-Black police discrimination decreased in Toronto over the past 25 years?

2. Have racial disparities in police stop, question and search activities (SQS) declined over the past 25 years?

3. Can racial disparities in police stop, question, and search activities be explained by other theoretically relevant variables?

Methodology

Data from three general population surveys are used in the following analysis. The first survey was conducted in 1994. Replications of this initial survey were conducted in both 2007 and 2019 [3]. The 1994 survey was conducted in 1994 on behalf of the Commission on Systemic Racism in the Ontario Criminal Justice System. The survey was conducted by York University's Institute for Social Research. It used random digit dialing techniques (RDD) to produce a random sample of over 1,200 Toronto residents who self-identified as either White, Black, or Asian (over 400 respondents from each racial group). This survey, the first of its kind in Canada, asked respondents detailed questions about their experiences with and perceptions of the Ontario criminal justice system. The 1994 survey was replicated in 2007, by the Hitachi Survey Research Center at the University of Toronto. This survey used RDD techniques to produce a random sample of over 1,500 respondents (500 from each racial group). Both the 1994 and 2007 surveys have resulted in several reports and peer reviewed articles (see Commission on Systemic Racism, 1995; Wortley, 1996; Wortley et al., 1997; Wortley and Tanner, 2003; Wortley and Tanner, 2005; Wortley and Owusu-Bempah, 2009; Wortley and Owusu-Bempah, 2011; Owusu-Bempah and Wortley, 2014; Wortley and Owusu-Bempah, 2016).

The 2019 survey was designed by the authors and conducted by Environics Research. While the first two surveys relied on random digit dialing technology, the 2019 survey used an online methodology. Importantly, the survey questions asked in 2019 were identical to the questions asked in both 1994 and 2007. The final sample consists of respondents from the Toronto region, 18 years of age or over. Quotas were set to ensure that we interviewed at least 450 respondents from each of three racial groups: 450 of the respondents self-identified as Black, 450 as Asian (including people of Chinese, Korean, Japanese backgrounds) and 550 as White/Caucasian. A total of 4,651 panelists received the survey invitation and were willing to complete the survey. Of those, 2,958 were terminated because they were either ‘over quota” or did not qualify for the survey (e.g. they lived outside of Toronto, were under 18 years of age, etc.). While 1,693 panelists qualified for the survey, only 1,450 completed the task (completion rate = 85.6%). The survey was conducted between May 16 and July 29, 2019 – prior to the George Floyd murder and global protests regarding police racism.

The data suggest that all three surveys produced samples that are representative of the Black, White, and Asian populations of Toronto. For example, consistent with Census estimates from each period, Black and Asian respondents are younger and much more likely to be born outside of Canada than White respondents. Furthermore, compared to their White and Asian counterparts, Black respondents report lower levels of educational attainment, higher unemployment rates, lower household incomes, and are more likely to reside in poor, high-crime neighborhoods (for more details about the survey methodologies and sample descriptions for all three surveys see Wortley et al., 1997; Wortley and Owusu-Bempah, 2009; Wortley and Owusu-Bempah, 2011; Wortley et al., 2020).

Findings

Perceptions of anti-Black and anti-Asian police bias

As part of the 1994, 2007 and 2019 surveys, all respondents were asked the following question: “In your opinion, do the police treat Black people better, worse or the same as White people?” Figure 1 highlights the percent of respondents, from each survey, who perceive that the police treat Black people worse or much worse than White people. The results suggest that perceptions of anti-Black police bias have increased over the last 25 years – especially among White and Asian respondents. For example, in 1994, only 51% of White respondents believed that the police treat Black people worse than White people. This figure rose to 59% in 2007 and to 62% in 2019. Similarly, in 1994, 56% of Asian people perceived that the police treat Black people worse than White people. This figure rises to 70% by 2019. The data also reveal that, throughout this twenty-five-year study period, the vast majority Black respondents have perceived anti-Black police bias. Importantly, this perception has not diminished with time. Indeed, the percentage of Black respondents who report that the Toronto police treat Black people worse or much worse than White people rose slightly from 76% in 1994 to 82% in 2019.

Figure 2 highlights the percent of respondents who perceive that the police treat Asian people worse or much worse than White people across the three survey periods. Several interesting findings emerge. White respondents are far less likely to acknowledge anti-Asian bias than anti-Black bias. Furthermore, the percent of White respondents who perceive that the police treat Asian people worse than White people has remained unchanged across this 25-year period. By contrast, Black respondents have come to perceive less police bias against the Asian community. For example, in 1994, 51% of Black respondents felt that the police treat Chinese people worse than Black people. This figure drops to only 41% by 2019. On the other hand, the perception of anti-Asian police bias has steadily increased amongst Asian respondents. In 1994, less than half of Asian respondents (46%) believed that the police treat Asian people worse than White people. This figure rises to 58% in 2019.

Clearly, despite government and police efforts to reduce racial bias and strengthen police-community relations, our data reveal that public perceptions of anti-Black racism in policing increased over the past 25 years. In other words, in the Canadian context, police bias and discrimination are just as important issues today as in 1994. Is it possible that perceptions of bias have persisted or become more entrenched despite reform and actual improvements in the police treatment of minority populations? To address this question, we move from an analysis of public perceptions of police discrimination to an analysis of racial differences in self-reported police stops.

Racial disparities in police stops and searches

Racial profiling has been an issue in Toronto for decades (see Mosher, 1998; Maynard, 2017). Black people in Toronto have long argued that they are much more likely to be unfairly stopped, questioned, and searched by the police than people from other racial backgrounds (Wortley, 1996; Tator and Henry, 2006). Unfortunately, official data on police stops are unavailable. The Toronto Police Service (TPS), unlike major police services in the United States and United Kingdom, has never collected data on the race of civilians involved in pedestrian or traffic stops. However, the TPS has documented “street checks” since the 1970s. While street checks do not document all police stops, they do document noncriminal, civilian encounters that the police believe are important for “intelligence purposes.” In the past, information about street checks was never released to the public. However, following a hotly contested freedom of information request, the Toronto Star obtained information on 1.7 million street checks documented by the TPS between 2003 and 2008. Subsequent data requests captured information on two million additional street checks completed between 2008 and 2013.

Overall, the TPS documented over 3.7 million street checks between 2003 and 2013 – an average of 336,300 per year (see Rankin, 2010a; Rankin et al., 2014). By comparison, between 2004 and 2012, the NYPD documented 4,135,000 stop, question, and frisk (SQF) investigations – approximately 459,000 per year (Torres, 2015). During this period, controlling for population size, Toronto's annual street check rate (14,600 per 100,000) was almost three times greater than the NYPD's SQF rate (5,598 per 100,000). Furthermore, consistent with NYC's stop and frisk statistics, Toronto's street check data reveal profound racial disparities with respect to involuntary police contact (White and Fradella, 2016).

The TPS street check data contain various pieces of information including the reason for the police stop, the time and location of the stop, the civilian's name and home address, and basic demographic information including age, gender, and skin color. Police argue that this type of information may help with future investigations by identifying potential suspects, victims, and witnesses. Critics, however, argue that street checks provide insight into police surveillance practices and reflect the types of neighborhoods and civilians that come under enhanced police scrutiny (Tator and Henry, 2006).

The following results stem from a re-analysis of TPS street check data, compiled from 2008 to 2013, conducted for the purposes of this paper. Only those cases in which the race of the civilian was recorded by the officer are included in the current analysis (sample size = 1,846,930) [4].

The data indicate that 25% of all street checks completed by the TPS between 2008 and November 2013 involved individuals described as “Black.” Census projections, however, suggest that only 8.08% of Toronto's population at the time self-identified as Black. In other words, Black people are 3.09 times more likely to appear in street check statistics than their representation in the Toronto population would predict (see Table 1). Further analysis reveals that during this period, the street check rate for Black people was 2,123.0 per 1,000. In other words, between 2008 and 2013, the TPS documented approximately 2.1 stops for every Black person in the city. By contrast, the street check rate for White people was only 653.7 per 1,000. Overall, the Black civilian street check rate is 3.25 times greater than the White rate.

In the wake of the racial profiling controversy, on January 1, 2017, the Government of Ontario implemented new regulations designed to govern how the police conduct street checks (Ontario Regulation 58/16). It appears that this regulation has virtually eliminated street checks documented by the Toronto Police Service (Tulloch, 2018). Figure 2 documents the number of street checks formally documented by the TPS between 2008 and 2019. The annual number of street checks documented by the TPS rose gradually from 323,041 in 2008 to a high of 403,662 in 2012. By 2012, street checks had become a public issue and the TPS was facing allegations of racial profiling. After an internal review of street check practices, and the adoption of a new street check policy, the number of street checks documented by the TPS plummeted to 189,536 in 2013 and to 24,364 in 2014 (PACER, 2014). A moratorium was put on street checks in 2015 and 2016, and no street checks were formally recorded by the TPS during this two-year period. Ontario's street check regulation came into play in 2017. Since that time, the TPS has, according to official statistics, documented only 28 street checks: 25 in 2017, two in 2018, and only one street check in 2019. Thus, according to official police statistics, street checks are a thing of the past.

The disappearance of street checks from official police statistics leads to new research questions: Has the elimination of street checks solved the problem of police racial profiling? Has the regulation reduced racial disparities in involuntary police contact? Importantly, previous research strongly suggests that community members and police officials often have very different definitions of what constitutes a street check. While community members tend to define street checks as being stopped, questioned, and searched by the police, the police traditionally focus on a much narrower range of technical activities associated with the collection of information for intelligence purposes (see Wortley, 2019). Thus, although street checks appear to no longer exist, we must further explore whether racial disparities in police stop, question and search activities (SQS) persist. Analysis of our survey data suggest that despite street check regulations, and a dramatic decline in officially documented street check incidents, alarming racial differences remain with respect to police SQS practices (see Figure 3).

Table 2 highlights the percentage of respondents who report being stopped by the police, during the past two years, for each year our survey has been conducted. Two important findings emerge. First of all, across all surveys, Black respondents report a much higher frequency of involuntary police contact than respondents from other racial groups. Secondly, the frequency of police stop activity increased significantly between 1994 and 2019. For example, in 1994, only 16.8% of Black respondents indicated that they had been stopped by the police on two or more occasions in the past two years. This figure rises to 21.0% in 2007 and 26.2% in 2019. Similarly, in 1994, only 4.7% of Asian respondents indicated that they had been stopped by the police on two or more occasions, compared to 12.5% in 2007 and 11.8% in 2019. By contrast, the stop rate for White people has remained relatively constant. In other words, according to these survey results, racial disparities in police stop activities have become more pronounced over this 25-year period.

These findings are particularly important in light of Ontario's new Street Check Regulation (O.Reg. 58/16). Although official statistics suggest that street checks were eliminated after the implementation of this law, the results of the 2019 survey, conducted more than two years after the street check regulation was imposed, suggest that Toronto-area police continue to stop and question civilians at a high rate. Furthermore, Black people continue to be stopped and questioned by the police at a much higher rate than people from other racial groups. Thus, although the Ontario Street Check Regulation may have eliminated the formal documentation of street checks, it has not decreased racial disparities in police stop and question activities. Eliminating the street check paper trail has not eliminated evidence of racial profiling. This finding also supports the need for mandating the collection of information on all police stops – not just those that result in a formal street check. We will return to this discussion in the final section of this report. In the next section we provide a multivariate analysis examining whether Black racial background remains a significant predictor of police stop and search activities after controlling for other relevant variables.

Predicting police stops

Using data from the 2019 survey, Table 3 presents a series of logistic regression models predicting the likelihood of experiencing police stops. These models statistically control for variables, including race, that the research literature suggests could be related to the probability of police contact (for a review of the literature on relevant variables see Wortley and Tanner, 2005; Wortley and Owusu-Bempah, 2011) [5].

To begin, we control for both Black and Asian racial background. White respondents are the default comparison group left out of the regression models. The regressions also control for respondent sex and age. The literature suggests that males and younger people attract more police attention than females or older people. The models also control for three social class measures – educational achievement, household income and community crime and disorder. The literature suggests that those with low levels of education and low incomes are more likely to reside in disadvantaged, high crime communities with a high police presence. Police surveillance practices in high crime communities should increase the chances of being stopped and questioned by the police.

Previous research literature tells us little about the relationship between immigration status and police contact. However, since a high proportion of our samples are immigrants, we included a variable capturing whether a respondent was foreign born or born in Canada. We have also included variables measuring whether a person has been a victim of a violent crime or whether their friends or family members have been crime victims. Research suggests that there is a strong relationship between criminal victimization and criminal offending. Thus, victimization measures may provide a proxy measure of involvement in a deviant lifestyle which could, in turn, predict the likelihood of police contact. To further capture respondent illegal activity, we also include variables that measure both illegal drug use and criminal record. We hypothesize that those who use illegal drugs and those that have previously been arrested for criminal activity will attract more police attention. Finally, we include four variables that measure the respondents' frequency of involvement in four different types of public activity: time spent driving in a motor vehicle, time spent walking or hanging out in public spaces, time spent shopping or hanging out in malls, and time spent engaging in late night party activities (i.e. the frequency of going out at night to attend bars, nightclubs, private parties, etc.). From the perspective of Routine activities Theory, the greater a respondent's involvement in public vs private activities, the greater their chances of being stopped, questioned, and searched by the police.

The results of the logistic regression analyses (see Table 3) reveal that Black racial background remains a strong predictor of police stops – even after other theoretically relevant variables have been taken into statistical account. In other words, racial differences in age, social class, residence in high crime communities, illegal drug use, criminal record, and routine activities cannot explain why Black people are more likely to be stopped by the police than people from other racial backgrounds. By contrast, the data suggest that, after controlling for other relevant variables, Asian respondents are no more likely to be stopped by the police than their White counterparts.

The data also reveal that Black people are particularly vulnerable to multiple police stops. After controlling for other variables, the logistic regression results suggest that Black respondents are twice as likely to experience one or more police stops, three times more likely to experience two or more police stops, seven times more likely to experience three or more police stops, and nine times more likely to experience four or more police stops.

In addition to Black racial identity, the results suggest that, in general, men are more likely to be stopped by the police than women. There is also a negative relationship between age and police contact. In other words, younger people are more likely to be stopped and questioned by the police, in the past two years, than older people. This is consistent with initial expectations. The results also reveal a strong relationship between community crime and disorder and the likelihood of experiencing police stops. In other words, consistent with our original hypotheses, people who reside in high crime communities are more likely to be stopped and questioned by the police than people who reside in low crime communities. However, it is important to stress that residence in high crime communities does not explain why Black people experience higher rates of police contact than White people [6]. Not surprisingly, crime victims report more police contact than non-victims, as do people who use illegal drugs and those with a criminal record. In terms of routine activities, respondents who drive frequently and those who often engage in late night party activities are more likely to report that they have been stopped and questioned by the police in the past two years. By contrast, time spent shopping or in other public venues does not predict police contact. These findings suggest that the more time spent in a motor vehicle the more likely you are to experience at least one traffic stop. The results also suggest that those who are frequently out in public, late at night, are especially vulnerable to police attention. Importantly, none of these factors explain the over-representation of Black respondents in self-reported police stops. The data suggest that in 2019 – when it comes to involuntary police contact – Black racial identity still matters.

Discussion

The results discussed above are somewhat pessimistic. Despite decades of protest, numerous government inquiries, and repeated promises of meaningful reform, our analysis reveals that the vast majority of Black people in Toronto still perceive discriminatory police treatment. These perceptions of racial bias are strongly supported by the fact that racial disparities in self-reported police stops have increased – not decreased – over the past twenty-five years. Other research reveals that racial disparities in police activity are not limited to police stop, question and search tactics. Indeed, recent studies, conducted on behalf of the Ontario Human Rights Commission, document that Black Toronto residents are much more likely to be subject to police use of force (Wortley et al., 2020) and discretionary arrest decisions (for example, in obstruct justice and disturb the peace charges) (Wortley and Jung, 2020) than people from other racial backgrounds. Together, these disturbing findings lend support to the argument that Canadian police-directed reforms, designed to address the issue of racial bias, are often performative rather than substantive (Worden and Mclean, 2017; Headley and Wright, 2019).

There is, however, reason for cautious optimism. To begin with, our data suggest that White and Asian residents of Toronto are becoming more aware of anti-Black racism within policing. The proportion of White and Asian respondents reporting that the police treat Black people worse than White people increased dramatically between 1994 and 2019. This trend is further supported by the fact that recent protests against police racism in Canada are becoming more racially diverse. While demonstrations in the 1980 and 1990s mainly involved Black people, protests that emerged after the 2020 slaying of George Floyd involved citizens of various racial backgrounds (Armstrong and Jacob, 2020). Hopefully, such allyship will serve to highlight the growing social importance of this issue and produce the political will needed to enact more meaningful reform.

A second positive development involves the collection of race-based data. Although academics and community members have long called for the dissemination of race-based data within the criminal justice system, Canadian police services have largely resisted these demands by enacting formal bans on the collection of racial information (Owusu-Bempah and Wortley, 2014). These bans have prevented both the identification of racial disparities and a proper evaluation of anti-racism initiatives. However, in 2020, the Toronto Police Service and Ontario Government announced plans to begin the collection of race-based data within policing and the broader justice system (Rankin and Gillis, 2019). We are hopeful that improved race-based data collection will lead to a better understanding of the role race plays in Canadian policing. Such data is also needed to establish benchmarks [7], evaluate efforts to reduce racial discrimination, and document public opinion with respect to law enforcement. At long last, Canada may finally have the data needed to properly assess police reform efforts and ensure police accountability (Robinson, 2020; Headley and Wright, 2019).

The findings presented above contribute to a growing body of literature documenting the global impact of police bias on Black and other racialized communities. While most published research has focused on the American context (Owusu-Bempah, 2017), our research reveals that racially biased police practices, and subsequent perceptions of police discrimination, are major problems in Canada and other Western democracies (Weber and Bowling, 2014; Owusu-Bempah and Gabbidon, 2020). Furthermore, as in the United States, our data suggest that Canadian police reform efforts have yet to improve public trust and confidence in law enforcement. In both Canada and the US, public perceptions of and experiences with police discrimination appear just as prevalent today as they were fifty years ago (see Hagan and Albonetti, 1982; Hagan et al., 2005; Henry and Franklin, 2019; Chenone et al., 2020; Graham et al., 2020). Clearly there is still much work to be done. This research laments the need for police agencies to further examine racial disparities in their stop and search practices and to appreciate the impact that police stops have on influencing public perceptions of the police (see also Wortley and Owusu-Bempah, 2011).

Figures

Figure 1

Percent of respondents who believe that the police treat Black people worse than white people, by respondent race and year of survey (Racial differences in response to questions about racial bias are highly statistically significant across all survey years)

Figure 2

Percent of respondents who believe that the police treat Chinese people worse than white people, by respondent race and year of survey

Figure 3

Number of street checks documented by the Toronto Police Service, 2008 to 2019

Table 1

Toronto Police Service street check data, by race, 2008 to 2013

Racial groups	Population size	% of population	Total number of street checks	% of street checks	Odds ratio	Street check rate (per 1,000)
White	1,454,030	54.09	950,457	51.46	0.95	653.7
Black	217,360	8.08	461,468	25.00	3.09	2,123.0
Brown	337,512	12.55	308,809	16.72	1.33	914.9
Other	679,840	25.28	126,196	6.83	0.27	185.6
Total	2,668,742	100.00	1,846,930	100.00	1.00	692.1

Table 2

Percentage of respondents who report being stopped by the police in the past two years, by respondent race and year of survey

Number of stops	Black			White			Asian
	1994	2007	2019	1994	2007	2019	1994	2007	2019
None	71.9	66.1	59.6	81.8	78.8	75.3	85.4	71.9	75.1
One	11.3	12.9	14.2	10.2	13.9	14.9	9.9	15.6	13.1
Two or more	16.8	21.0	26.2	8.0	7.3	9.8	4.7	12.5	11.8
N	417	513	450	435	505	550	405	504	450

Table 3

Logistic regression models predicting number of police stops experienced by respondents over the previous two years

Independent variables	One or more police stops		Two or more police stops		Three or more police stops		Four or more police stops
	B	Odds	B	Odds	B	Odds	B	Odds
Black	0.654***	1.923	1.057***	2.879	1.983***	7.265	2.202***	9.039
Asian	0.072	1.074	0.247	1.280	0.690	1.994	0.847	2.332
Age	−0.017***	0.983	−0.023***	0.977	−0.020*	0.981	−0.016	0.984
Sex (1 = male)	0.427**	1.533	0.549**	1.732	0.662**	1.939	0.520	1.682
Education	−0.052	0.950	0.009	1.009	0.052	1.053	0.108	1.114
Income	0.047*	1.048	0.001	1.001	−0.004	0.996	−0.006	0.994
Foreign Born	−0.098	0.906	−0.201	0.818	−0.281	0.755	−0.223	0.800
Community Disorder	0.088***	1.091	0.099***	1.104	0.119***	1.126	0.133***	1.142
Victim of Violence	0.436*	1.547	0.306	1.358	0.615*	1.849	0.814**	2.256
Friends/Family Victims	0.172	1.187	0.302	1.352	0.472	1.603	0.422	1.525
Illegal Drug Use	0.259	1.296	0.874**	2.397	0.350	1.419	0.264	1.302
Criminal Record	0.547*	1.728	0.170	1.185	−0.255	0.775	−0.127	0.881
Driving Frequency	0.109*	1.115	0.019	1.019	−0.087	0.917	−0.078	0.925
Time in Public Spaces	−0.060	0.942	−0.003	0.997	0.033	1.034	0.046	1.047
Time shopping	0.037	1.038	0.055	1.057	0.030	1.031	0.033	1.033
Late night party activities	0.335***	1.397	0.302***	1.353	0.461***	1.586	0.531***	1.701
Constant	−2.722***	0.066	−3.372	0.034	−5.166	0.006	−6.667***	0.001
Nagelkerke R-Square	0.200		0.243		0.341		0.360

Note(s): ***p > 0.001; **p > 0.01; *p > 0.05

Notes

1. The authors use the term performative work here to suggest that the efforts taken by the Toronto Police Service were intended to give the impression that the service was working to address racial discrimination without taking steps necessary to actually reduce or eliminate racial discrimination. Put another way, these efforts could be described as window dressing.

2. In the vast majority of cases, contact cards were not filled out during police encounters that ended in arrest or criminal charges. In such cases, a record of arrest and/or criminal incident report is used to capture relevant information. Street checks or contact cards, on the other hand, were typically filled out in cases where criminal charges were not laid but where the police officer still wanted to record personal information about the civilian stopped and details about the encounter. The TPS argued that this information was valuable with respect to identifying future victims, offenders and witnesses.

3. The lead author (Wortley) developed and implemented the 1994 survey on behalf of the Commission on Systemic Racism in the Ontario Criminal Justice System. Wortley secured a SSHRCC research grant to develop and implement the 2007 survey. Dr Owusu-Bempah was also a researcher on the 2007 project. Both Owusu-Bempah and Wortley were involved with securing funding for the 2019 project and implementing the survey.

4. Between 2008 and November 2013, the Toronto Police Service completed 2,026,258 contact cards or field information reports. However, information on the race of civilian was missing in 179,328 cases (about 9% of the sample). These cases are left out of the current analysis.

5. A previous version of the paper included a methodological appendix that fully described the coding of all dependent and independent variables used in the analysis. Due to space constraints, the editors asked the authors to delete this appendix. For a copy of this appendix, readers are asked to contact Dr Scot Wortley, Center for Criminology, University of Toronto, at scot.wortley@utoronto.ca

6. Black respondents scored significantly higher on the Crime and Disorder Index (mean = 7.00) than white (mean = 5.81) and Asian (mean-5.56) respondents (F = 14.105; df = 2; p > 0.001).

7. In this context a benchmark is a reference group upon which data can be compared. For example, the percentage of Black people represented in police stop data, compared with the percentage of Black people in the general population or the driving population on a given highway available for the police to stop. Benchmarks may be internal or external. For further detail, see Smith et al. (2021).