Racial Profiling in Policing

Does Racial Profiling Occur in the British Police?

ABSTRACT

A controversial issue in policing today is the alleged use of racial profiling by police to stop/search citizens. Currently only a small amount of research exists concerning this practice, none of which has been conducted in the UK. This article summarises the major cases that established the existence of racial profiling in the US, identifying with the strengths and weaknesses of the methods used while accounting for the problem that racial profiling characteristics have yet to be identified and conceptualised.

Using data collected from the stop/searches of three different actors, a White male, Asian male and a Black male in six different locations of demographic significance while controlling for suspects demeanour; each stop/search will be analysed to conclude whether racial profiling takes place by the British Police. It can be presupposed that in line with previous police research in both the UK and the US, the actors belonging to the ethnic minority groups, will suffer a higher number of stop/searches.

Does Racial Profiling occur in the British Police?

The purpose of the proposed research is to explore whether racial profiling take places by police officers when stop/searching citizens within the UK. Recent Home Office reports and studies have revealed minority ethnic groups are subjected to heightened rates of stop/search in relation to the ethnic composition of the resident nation. Figures for 2003/2004 showed that the rate of stop/search for black people was nearly six and-a-half times that for whites, while that for Asian people was nearly twice that for whites (Home Office). The proposed research will be carried out using officers on foot patrol in Manchester, a large diverse community where heightened stop/search rates amongst ethnic minorities are apparent (Greater Manchester Police Statistics 2006).

Although current literature exists concerning this problem (Harris, 1999; Weitzer & Tuch, 2002), it has only been reviewed with regards to the US police service where racial profiling was found to take place (Lamberth, 1997). No such literature or studies have been carried out within the UK.

Therefore by researching whether police officers engage in racial profiling when stop/searching citizens, individual officers and their resultant Constabulary will become accountable for the injustice practice and can be disciplined as required. Furthermore the reasoning behind such stop/searches may become apparent. This may in addition, allow for the public opinion and trust of the police to become more favourable, than recent opinion polls have suggested (FitzGerald et al. 2002) and thus allow for a more effective police service (Skogan, 2004).

Background to the Study

One of the most prominent issues facing the police today is the use of race as a criterion in police decision making during discretionary traffic and field stops, often described as racial profiling (Engel, Calnon, & Bernard, 2002). This is especially the case in minority and ethnic communities where police stops have long been viewed as controversial (Weitzer & Tuch, 2002). Although there is a significant literature on the relationship between race, police, and the administration of law (see Chambliss, 1994), recent literature shifts the attention toward the perceived racial profiling problem (see Harris, 1999; Russell, 2001; Weitzer & Tuch, 2002) and methodological discussions concerning measurement difficulties when identifying racial profiling (Ramirez, McDevitt,&Farrell, 2000; Walker, 2001).

Racial profiling is often referred to, as the alleged law enforcement practice of using skin colour as a pretext to stop, question, or search minorities, and can occur during any situation in which minorities are stopped by law enforcement officials (Harris, 1999; Russell, 1999; Smith &Petrocelli, 2001; Walker, 2001).Weitzer and Tuch (2002) recently referred to racial profiling as the “use of race as a key factor in police decisions to stop and interrogate citizens” (p. 435).

Although some form of profiling in law enforcement has long been a practice employed by police (Piliavin & Briar, 1964), it is only over the past few years that it has become a controversial topic in political, academic, and social arenas (Russell, 2001). At the core of the racial profiling issue is the supposition that police target minorities, particularly Blacks, during their normal patrol duties with the belief that minorities are more likely to be guilty of having committed a crime than have Whites (Walker, 2001).

Racial Profiling has increasingly become a topic under scrutiny throughout the democratic world, and it is difficult to overstate the importance of clear conceptualization in this research proposal. A firm understanding of the meaning and parameters of key concepts is necessary for identifying valid and reliable measures and adequately operationalising variables. In the literature to date, there appear to be at least two clearly distinguishable definitions of the term “racial profiling”, a narrow definition and a broad definition.

For the purpose of this research a broad definition of racial profiling will be used; ‘racial profiling occurs when a law enforcement officer uses race or ethnicity as one of several factors in deciding to stop, question, arrest, and/or search someone.’ Under the narrow definition race must be the sole issue in deciding to stop/search someone. Stop/search is primarily an investigative power used for the purposes of crime detection or prevention in relation to a specific individual at a specific time (Lustgarten, 2002).

In the UK a stop occurs when a person is asked to account for their actions or movements for investigation purposes. Under section 1 of the Police and Criminal Evidence Act 1984, a constable does not have the right to stop and search ‘a person or vehicle or anything in or on a vehicle unless he has reasonable grounds’ for suspicion that stolen property or prohibited articles are being carried. In 2002/2003, there were 869,164 stops and searches under this legislation (Home Office). Amongst them, 118,548 were Black people (14%), 58,831 were Asian people (7%) and 11,468 were of other minority ethnic origin (1%) even though Black and minority ethnic represents only around 7% of the whole population; noticeably the disproportionate use of stop searches on ethnic minorities has become apparent.

Although literature surrounding the problems of stop/searches in the UK exists (Willis, 1983; Smith, 1983; Elliott and Quinn, 2002) no research has been carried out to investigate whether the problems in the heightened number of stop/searches of ethnic minorities is in fact, due to racial profiling taking place by the police. Indeed research suggesting police racism has been carried out (Institute of Race Relations, 1987; Bowling and Phillips, 2002), the most damming of which is the Macpherson Report 1999.

This report identified “Institutional racism and a failure of leadership by senior police officers” (Macpherson 1999, para 46.1). Macpherson highlighted a general lack of trust and confidence in the police amongst ethnic minorities and noted that the experience of black people over the last 30 years has been that ‘we have been over policed and to a large extent under protected’. It concluded that the over-representation of racial minorities in the national stop and search figures led to the ‘clear core conclusion of racist stereotyping’ (Macpherson 1999, para. 6.45).

Evidence of subcultural patterns of racism and machismo began to be uncovered within the British police service during the early 1970s against a background of a racialised moral panic about ‘mugging’ (Hall et al. 1978; Rowe 2004). A submission to the Royal Commission on Criminal Procedure by the Institute of Race Relations (1979, 1987) drew attention to the mass stop and search of black people.

This was further publicised due to the Brixton riots in 1981. Lord Scarman criticised the heavy-handed approach to policing in Brixton and highlighted the role of operation ‘‘Swamp 81”, which involved more than 120 officers patrolling the area with the instruction to stop and search anyone that looked ‘‘suspicious”. Over 4 days, 943 people were stopped and 118 were arrested, more than half of whom were black (Bowling and Philips 2002).

Reiner’s influential analysis of the police occupational culture (Reiner, 1985) drew on earlier work by Skolnick (1975), to provide a description of what he called the ‘core characteristics of cop culture’. These included racial prejudice, some of which Reiner links to the dominant attitude of the majority towards minorities, and some to the transactional relationships between police officers and the minorities they police against.

Thus, it is apparent that racial prejudice has been identified to exist within the British Police Force, however no studies have identified whether the stop/search of these individuals was down to racial profiling by officers. One study suggested that, relative to the resident population, Asians are eighteen times and African-Caribbean’ twenty-seven times more likely to be stopped by the police than whites (Bowling and Phillips 2002). Analysis of Home Office figures suggest that between 1999 and 2003, relative to the resident population, blacks are eight times and Asians two-and-a-half times more likely than whites to be stopped by the police. The same report suggests that in the two years prior to 2003, the number of stops of whites fell 19% while stops of black and Asians each increased by 28% (Institute for Race Relation 2003).

Yet the reasoning behind these disproportional figures has never been properly investigated. The Macpherson report (1999) attributed disproportionate levels of stop and search to “stereotyping”, police racism may be ‘unwitting’, but nonetheless is an expression of the subjectivity of officers, albeit culturally shared. For instance, it is suggested that police are generally more suspicious of racial minorities than of white people (FitzGerald and Sibbitt 1997). However, police attention may not be as misplaced as might be supposed. It has long been observed that the ‘strike rate’- the number of arrests arising from stop/searches is approximately equal across racial categories. Stop/searches only lead to arrest, in only 13% of cases in 2003/2004 (Home Office 2005).

In the US where research is more extensive and racial profiling has been identified as police practice; Weitzer (1999, 2000) and Weitzer and Tuch (1999, 2002) investigated the public’s perceptions of racial profiling. Their work has moved what we know about perceptions on profiling from anecdotal to objective. Most recently, Weitzer and Tuch (2002) reported results from a 1999 Gallup poll that indicates attitudes toward the police in general and profiling in particular.

In addition, their work suggests that individuals who believed that they had been a victim of racial profiling expressed dissatisfaction with the police in general and reported that profiling was a widespread tactic. In the UK, the “Policing for London” report found that ‘being stopped on foot’ emerged as the prime factor amongst a range of ‘police contacts and experiences of crime’ associated with negative views of the police (Fitzgerald et al. 2002). In general however, research has identified favourable opinions towards the police (Brandl, Frank, Worden&Bynum, 1994) most indicates difference of opinions along racial and ethical lines (Barkan&Cohn, 1998). Surveys by Flanagan and Vaughn (1995), Huang and Vaughn (1996), and Williams, Thomas, and Singh (1983) all show strong racial differentiation in opinions towards police (cited in Barkan & Cohn, 1998).

While it is important to review literature supporting my proposed research, it is also fundamental to note Jefferson (1993), who persuasively argues that race is only one, relatively minor component explaining why ‘criminalization’ (stop/search, and arrest) falls more heavily upon particular sections of the population. Other components and rationale as to the high number of stop/searches of ethnic minorities will be considered as a matter of course.

To date, racial profiling is frequently described as a serious problem that occurs all too often (Cleary, 2000). A Gallup poll conducted in the US reports that 56% of Whites and 77% of Blacks believe that racial profiling is widespread (Newport, 1999 cited in Weitzer & Tuch2002). However, the true extent of racial profiling in society is unknown as little research has been done and demographic information on stops has not routinely collected by police (Cleary, 2000; Harris, 1999b; Ramirez et al. 2000; Walker, 2001). Furthermore, of the research that has been done, estimates are often difficult to interpret and comparisons must be viewed with caution because of differences in the conceptualization and operationalisation of racial profiling.

In the US, where previous studies have been conducted to assess the presence or extent of race-based policing (see Ramirez et al. 2000). The most common has been to compare the proportional representation of racial and ethnic groups in a certain population against the proportional representation of racial and ethnic groups stopped, searched, and/or arrested by police officers within the same population. Within this methodological approach there are at least two data collection strategies.

The first, relies on police officers to report the race and ethnicity of the people they stop/search in an information-gathering mechanism independent of, and in addition to, the routine information-gathering procedures (e.g., citations or activity reports) currently supported or required by their departments. This requires the officer to complete an additional form subsequent to each stop/search (Cordner et al. 2001; Norris et al., 1992; Smith & Petrocelli, 2001). They then compare the racial and ethnic proportions of the individuals stopped/searched against the proportional representation of racial and ethnic groups that existed within the population at the same time (Cox et al. 2001; Meehan&Ponder, 2002; Zingraff et al. 2000). Some of these studies have led to the identification of racial profiling in practice within US police departments.

A 1999 report by David Harris, “Driving While Black: Racial Profiling on Our Nation’s Highways” cites numerous accounts of disparate treatment toward minorities by police. One anecdote used was that of Elmo Randolph, an African American dentist, who testified that over a 3-year period, he was stopped 50 times on the New Jersey Turnpike, and asked if he was carrying drugs, and released. None of the stops resulted in the issuance of a citation for any motor vehicle infraction (Kocieniewski, 1999a). This evidence certainly suggests racial profiling to have taken place on numerous occasions against a single member of the population due to his appearance.

Another finding by Harris was that police stop people of colour travelling through predominately White areas because the police believe that people of colour do not “belong” in certain neighbourhoods and may be engaged in criminal activity. This type of profiling was reported by Alvin Penn, the African-American deputy president of the Connecticut State Senate. In 1996, a Trumbull, Connecticut, police officer stopped Penn as he drove his van through this predominately White suburban town.

After reviewing Penn’s license and registration, the officer asked Penn if he knew which town he was in (Bridgeport, the state’s largest city, where Blacks and Latinos comprise 75 percent of the population, borders Trumbull, which is 98 percent White). Harris (1999b) as a result argued that racial profiling was institutionalized in US law enforcement in the mid-1980s through Operation Pipeline, a Drug Enforcement Agency training program on using traffic violations as pretexts for stopping persons who fit drug courier profiles, especially on roads known as drug “pipelines.”

One officer testified that he focused on “rented cars from south Florida, driven by blacks or Latinos” (Webb, 1999, p. 126) which certainly suggests a racial bias but is nevertheless true to the original Operation Pipeline drug courier profile. Although this small sample of anecdotal evidence does not prove that police officers actively engage in racial profiling, it is representative of thousands of personal stories catalogued throughout the US.

As a result of the growing body of individual accounts of racial profiling, scholars began examining the relationship between police stop/search practices and racial characteristics of individual drivers using empirical data. The majority of empirical research collected to date has been used in expert testimony accompanying lawsuits. Wilkins v. Maryland State Police(1993) was one of the first cases to introduce empirical evidence of racial profiling into the court record. Dr. John Lamberth, a professor of psychology at Temple University, conducted an analysis of police searches along I–95 in Maryland.

The research took place following a Maryland trooper stopping Robert Wilkins’s rented car as the family returned from a funeral in Chicago. Over Wilkins’s objections, the trooper ordered occupants out of the car, to stand on the side of the highway for an extended period of time, at night, in the rain, until the arrival of a unit with a drug-sniffing dog. No drugs or violations had taken place an as a result Wilkins sued the Maryland State Police for violation of civil rights (State of New Jersey v.Pedro Soto, et al. 1996 cited in).

Lamberth (1996) analyzed Maryland State Police stop/search data as part of the settlement agreement, comparing those findings to another rolling survey of the I-95 highway user and law-violators populations. Using data released by Maryland State Police pursuant to the settlement, Lamberth compared the population of people searched and arrested with those violating traffic laws on Maryland highways. Race could be determined for 5,555 of the 5,741 driver’s observed (96.8%), and 5,354 drivers (93%) were observed violating motor vehicle laws (Lamberth, 1996).

Lamberth’s definition of “violation” is controversial but has a supportable rationale. His assistants drove at the posted speed limit (55 miles per hour or 65 miles per hour, in different areas), counting the drivers who passed them as “violators”. His violator survey indicated that 74.7 percent of speeders were White, while 17.5 percent were Black. In contrast, according to State Police data, Blacks constituted 79.2 percent of the drivers searched; this proposed that Blacks were 4.85 times as likely as Whites to be pulled over by New Jersey troopers. Lamberth concluded that the data revealed “dramatic and highly statistically significant disparities between the percentage of Black I–95 motorists legitimately subject to stop by the Maryland State Police and the percentage of Black motorists detained and searched by troopers on this roadway.”

A further study by Spitzer(1999) gauged stop/frisk rates according to the racial breakdown in New York City. To accomplish this task, researchers analyzed the UF-250 forms from the NYPD. These forms are detailed reports that every officer must complete for each stop made during their shifts.

The reports include a detailed description of the stop, incorporating characteristics such as the race of the person stopped, the reason he or she was stopped, the time the stop was made, where the stop occurred, and who the officer was that made the stop. Analysis of the UF-250s between January 1, 1998 and April 1, 1999 showed that Blacks comprised 50.6% and Hispanics 33.0% of all those stopped or frisked, although they only represented 25.6% and 23.7% of the New York City population, respectively. Conversely, Whites, who comprise 43.4% of the New York City population only represented 12% of those stopped or frisked. Hence, Blacks were 6 times more likely and Hispanics 4 times more likely to be stopped or frisked in New York City than were Whites. Further research at the precinct level showed that, of the 75 precincts that form the NYPD, only 1 of the 10 precincts with the highest stop rates had a “majority” of White residents, whereas all but 2 were in the lowest 10 precincts (for stop rates) (Spitzer, 1999).

The Spitzer (1999) report also gauged whether stop/frisk differences would mediate after controlling for the crime rate according to race. The data revealed that after holding the race-specific precinct crime rates and precinct population composition by race constant, the Black stop/arrest ratio (1.54) was 23% more than the White ratio (1.24). The Hispanic stop/ arrest ratio of 1.72 was 39% more than the White stop/arrest ratio. The researchers reported that the crime rate in New York City could not substantially account for the disparities in stop rates between Blacks, Hispanics, and Whites (Spitzer, 1999). Although this study differed in method from previous research conducted it still produced the same outcome suggesting racial profiling occurred within the police departments.

However, despite empirical evidence proposing racial profiling by police officers, statewide studies in Connecticut, Texas, and Washington all failed to produce consistent findings of substantial and systematic disparity of stops with respect to the race or ethnicity of the person (Cox, Pease, Miller, & Tyson, 2001; Texas Department of Public Safety, 2000; Washington State Patrol , 2001).

In a study of the Richmond, Virginia Police Department, Smith and Petrocelli (2001) found that in a comparison of the percentages of those being stopped to that of the driving eligible population, Blacks were 46% more likely to be stopped than Whites. They reported that of the 2,673 motorists stopped, Black motorists comprised over two thirds while only making up 52% of the driving eligible population. Similar to the 1997 Lamberth study, the Richmond Police searched minorities 71.0% of the time compared to only 28.9% of Whites. Petrocelli, Piquero, and Smith (2003) also used the Richmond data to examine racial profiling at the census-tract level by studying the factors related to police stop/searches.

Their analysis indicated that the area crime rate was the only significant factor influencing police stop patterns. In other words, police stops were primarily a function of the crime rates of particular patrolling areas in Richmond (Petrocelli et al., 2003). Home Office research in the UK had similar findings. Miller (2000) found a strong positive correlation between locations where stop and search was greatest and crime was highest. Walker’s (2001) study of the San Jose Police Departments’ data on traffic stops yielded similar results to prior research, suggesting race was the significant factor in determining those stop/searched.

As a result of previous literature suggesting racial profiling occurs within the US police force, I advocate that the research proposed will most certainly be of importance to the UK and future policing. Within the UK literature by Norris et al. (1992) found not only that young blacks were stopped very much more frequently than other racial groups, but that these stops were made on a more speculative basis. FitzGerald and Hough found that being black was a good predictor for being stopped on foot and in a car, along with being under 30 years old and male (FitzGerald et al.2002).

In 2002 research was commissioned by the Thames Valley Police. Conducted in Reading and Slough, the survey was part of a larger study that included in-depth interviews with police officers of different ranks plus analysis of police records of stop/searches that had occurred (detailed in Qureshi 2005). The reason for commissioning the research was official anxiety at the apparent disproportionality between the racial composition of the residential population and the profile of those stop/searched in these two police divisions. Fortuitously, the racial composition of these two towns exhibited vast differences.

Reading has a mixed ethnic profile that includes a substantial proportion of black people, whereas Slough has a distinctively large Asian minority; the differences between these two towns offered the opportunity to examine any differences in patterns of ‘availability’ and police stop/search practices. The research methods included direct observation of the ‘available population’, analysis of the stop/search database, and interviews with officers about recent stop and searches that they had conducted. When results were compared to the resident populations of each division (2001 Census), police stop and search figures certainly suggest significant disproportionality in both locations.

However, when we consider the ‘available population’, we find a rather different picture. White people are much less evident amongst those available to be stopped and searched in both towns, especially in Slough. Black and Asian people are marginally, but crucially, more abundant in Reading and far more numerous in Slough. The result’s show that, in Reading, the pattern of stop/search reflects pretty closely the available population, whilst in Slough, there is significant disproportionality, with white people being over-represented amongst those stopped/searched and Asians underrepresented.

Whilst the literature analysed suggests that racial profiling may take place, it is important to consider the alternative rationale behind the higher number of stops and the methodological problems in achieving the data to analyse, chief amongst them is the reliance upon official figures of stop/searches. There is good reason to believe that these figures underestimate the actual number of stop/searches conducted by police officers (Quinton et al. 2000; Smith and Gray 1983).

Given that the issue of racial profiling at its essence involves the use of police discretion, it’s important to recognize the potential role that characteristics of a particular neighbourhood or police district play in the decision to make a stop/search. In the UK the use of reasonable suspicion is often down to the discretion of the individual officer. Smith and Gray (1983) in a study proceeding PACE1984, examined the application of ‘reasonable suspicion’.

They observed that, when carrying out stop/searches, police officers rely predominantly on their own instincts, which could cause elements of race and class bias. It was Smith (1986) who positioned that in order to understand police behaviour, one must consider the environment in which police operate. The exercise of discretion occurs outside the supervisory purview of the police station (Bowling et al., 2004). This can provide the opportunity for police officers to exercise their discretionary powers based upon their own individual prejudices (Kleining, 1996).

This is apparent in research identifying police stop/search practices being driven by; age, class, gender and race (Skogan, 1990 and 1994; Clancy et al. 2001). Hence, racial profiling could be implied by officers due to their individual prejudices without being an actual ‘feature’ of the police service as a whole. It may also stem from supervisor guidelines and differences in training and (Verniero & Zoubek, 1999 cited in Reitzel & Piquero2006)).

Several studies have found the ethnic composition of the available population differs remarkably to that of the residential population, and that this difference may go a long way in accounting for the apparent disproportionate use of stop/searches against people from minority communities (Miller et al. 2000; Waddington et al. 2004). One interest in all the literature concerns the baseline (Smith et al., 2000; Smith& Petrocelli, 2001; Thacher, 2001; Walker, 2001).

Currently, there is no clear agreement in regard to what a proper baseline should be in measuring racial profiling (Smith & Petrocelli, 2001; Thacher, 2001). Zingraff et al. (2000) reported that using statewide data such as census figures or number of licensed drivers in the state or a particular region is problematic because it cannot account for the distribution of people in a locale at a specific time and it does not account for differences in patrolling. Measures used in previous studies also fail to account for non-residents travelling or residing in the area under study, which itself impacts baseline measures (Cox et al 2001).

One supposition behind the higher number of stop/searches of ethnic minorities is that different racial or ethnic groups place themselves at greater or lesser risk of being stopped by the police through their differential use of public space. FitzGerald speculates that young Asian men increasingly seek to escape the confines of the family home, finding refuge amongst their companions in public places.

They disproportionately occupy public spaces as groups that sometimes engage in the kind of rowdiness that attracts police attention and are vulnerable to being stop/searched and arrested for possession of small quantities of cannabis (FitzGerald 1999). Bowling and Phillips (2001) also identify the racialised nature of ‘availability’. Racial patterning of unemployment, homelessness and school exclusion may be reflected in the composition of the ‘available population’ and, even if police proportionately select from amongst this ‘available population’, it remains the case that racial minorities may be more exposed to stop/searches. Hence class not race may be a cause of the higher rates of stop/search (Don et al. 2004).

The use of the residential population as the comparative basis for these statistics is problematic, not least because different sections of the population may use public space differently. It has been noted that the elderly are increasingly disinclined to venture onto the streets as pedestrians because of their fear of crime (Chivite-Matthews and Maggs 2002). Women too seem dissuaded from using the streets, especially after dark (Gilchrist et al. 1998; Mirrlees-Black et al. 1996). Hence, the overwhelming disproportionality in stop/searches experienced by young men of all racial groups and ethnic identities may simply attest to their greater availability for being stop/searched (Home Office). Rather than any particular selectivity on the part of the police. It appears necessary, therefore, to compare stop/search figures, not with the racial composition of the resident population, but with the composition of those available for being stopped because they frequent public places.

While the literature clearly presents empirical evidence of stop/searches of ethnic minorities at a higher level in both the UK and the US, it fails to properly account for the reasoning behind these high rates or use an accurate baseline of population to compare the results against. As a result, it remains a difficult task to construct reliable estimates of the unique individual and area-specific factors that contribute to explanation of police decision making and racial profiling. Engel et al. (2002) reported that non-White suspects may be more disrespectful toward police, which could be linked with higher arrest rates for minorities, especially young Black males.

On a related note no published racial profiling study has included information on citizen demeanour, yet a longstanding body of literature points to the importance of suspect demeanour on police decision making (Klinger, 1994; Lundman, 1996; Worden, Shepard, &Mastrofski, 1996). The concept of demeanour as a result will also be analysed in the proposed research. This will allow a greater understanding of whether racial profiling occurs in the British Police Force or if the high level of stop/searches is down to other rationale.

The Study

In the process of conceptualising racial profiling, it is important to consider the unit of analysis and the possibility of effects at different levels of aggregation. A better understanding of the heightened stop/search rates of minorities in the UK will be achieved through careful attention to the process of stop/search by individu