Over the past two decades, the police have often been the focus of praise for their innovations in policing strategies and their leadership in rolling back what seemed an inevitable rise in crime rates. In one of the earliest examples of this positive recognition of the role of the police in fighting crime, William Bratton, then commissioner of the New York City Police Department, was pictured on the cover of Time, one of the most important U.S. news outlets at that time, with a headline: “Finally, We’re Winning the War against Crime. Here’s Why” (TIME, 1996). Such headlines were common at the beginning of the new millennium, and they continue to be common as police agencies take credit for controlling crime in American cities (see, e.g., Youmans, 2000; Wood, 2001; Allen, 2002; Rashbaum, 2003; Williams, 2003; Cella, 2004; Dowdy, 2004). They express a more general acceptance by the public that the police play a key role in doing something about the crime problem.
It is worth noting that this confidence in the ability of police to address crime is of recent vintage. The conventional wisdom, at least from the 1960s until the mid-1990s in the United States, was that police had very little impact on crime rates (Bayley, 1994; Gottfredson and Hirschi, 1990). The origins of this view can be found in the 1967 report of the President’s Commission on Law Enforcement and Administration of Justice, The Challenge of Crime in a Free Society, which detailed the relationship between so-called root causes and crime and raised questions regarding the practices common in U.S. policing (President’s Commission on Law Enforcement and Administration of Justice, 1967). It was reinforced by a series of academic studies that challenged the crime-control effectiveness of standard police
practices (see, e.g., Kelling et al., 1974; Levine, 1975; Spelman and Brown, 1984). If crime is rooted in poverty and deprivation, then what could police do to stop it? In turn, systematic study of the practices that dominated the efforts of the police to do something about crime did not yield positive results. It was thought that police should focus on other tasks such as bringing offenders to justice regardless of whether such work affected the crime rate, peacekeeping tasks such as intervening in domestic disputes, providing help and assistance to those in need by responding to emergency calls, and traffic control.
In part as a response to research that challenged the effectiveness of traditional policing strategies, the 1980s and 1990s saw the emergence of a series of innovative police practices. They could be contrasted with the “standard” models of policing in earlier decades by their focus on taking a proactive approach to crime problems. Most of the standard practices of policing simply reacted to the occurrence of a crime. They were part of the police role as first responders and agents responsible for bringing offenders to justice. The new strategies proposed by police and scholars were proactive, in that they went beyond the obligations of the police to respond to the occurrence of crime and to investigate and bring offenders to justice; instead, they focused on policing approaches that could be successful in crime prevention irrespective of whether they had been seen in the past as traditional components of police practice.
These innovative proactive policing strategies have now become part of the national lexicon. The growing perception that the police could prevent crime was buttressed by a National Research Council ([NRC] 2004) report, Fairness and Effectiveness in Policing: The Evidence. That report noted that research on the standard models of policing common in the United States at the time did not support claims for crime control. However, the report argued that evidence was beginning to emerge that promising new proactive policing strategies could prevent crime.
From the perspective of the police and police researchers, this was exciting news, but the evidence base reviewed in the 2004 NRC report was still developing and did not cover some important proactive approaches. A number of innovations in proactive policing were just beginning to be examined. While, for example, a series of randomized field experiments were found to support the effectiveness of hot spots policing, there was much less rigorous research at that time on problem-oriented policing, broken windows policing (which seeks to prevent serious crime by addressing disorder and minor offenses), and “pulling levers” or “focused deterrence” policing (which emphasizes identifying dangerous offenders and using multiple police and community pressures to reduce crime). Accordingly, while the 2004 NRC report provided a glimpse of the potential for proactive policing, the approaches and the research on their outcomes had only begun to
be developed. This current report was commissioned because it was time to take a fuller look at whether proactive policing can reduce crime and disorder and, as important, which of the strategies developed have the greatest promise for crime reduction.
But the crime-control effectiveness of proactive policing should not be examined without consideration of its broader impacts on law and the community. Democratic societies require that police balance the provision of public safety from crime with other important values, such as police adherence to law, economy in the use of coercion, the provision of service, and attentiveness to fairness and the general welfare of citizens in the community (Bayley, 2006; Bittner, 1970; Manning, 2010; Muir, 1977). News reports over the past few years focusing on conflicts between the police and the public are a reminder that policing exists in a complex set of social contexts and that effectiveness in reducing crime and disorder is not the sole metric by which policing strategies should be evaluated. Americans have been confronted by difficult images of police brutality and even killings by police (Baker, Goodman, and Mueller, 2015; Buchanan et al., 2015; Dewan and Oppel, 2015; Graham, 2016). High-profile incidents of fatal violence directed at police officers in New York City and Dallas, Texas, have been interpreted as a response to those events (Achenbach et al., 2016; Mueller and Baker, 2014). Protests, and in some cases rioting, throughout the nation have focused on policing and often on what are perceived as unfair and abusive police practices (Domonoske, 2016; Lee et al., 2016; Nolan and Chokshi, 2016; Payne, 2014; USA Today, 2016). In particular, Blacks and other non-White groups have expressed concerns about how they are treated by the police and about the differential impacts of policing in non-White communities. The emergence of the Black Lives Matter group during this period suggests the heightened concerns of specific non-White communities to the policies and practices of the police (see Appendix A). This heightened discontent with policing, in a way reminiscent of the 1960s and 1970s, stimulated a blue-ribbon presidential task force to call for increased attention to strengthening the bonds between the police and the community (President’s Task Force on 21st Century Policing, 2015).
The committee authoring this report was tasked with considering how proactive policing strategies bear upon these concerns. It is not enough simply to identify “what works” for reducing crime and disorder; it is also critical to consider how proactive policing affects the legality of policing, the evaluations of the police in communities, the potential abuses of police authority, and the equitable application of police services and police interference in the everyday lives of citizens.
Are the new proactive policing strategies the source of the growing challenge to the legitimacy of police in the United States? Some media commentators have made this connection directly. For example, Gloria Tso
(2016), in the Washington, D.C., newspaper The Hill, draws a direct connection between abusive and illegal police practices and hot spots policing, one of the proactive strategies that has emerged in the new millennium:
The epidemic of police brutality—primarily affecting black males—can be linked to the history of a technique called hot spot policing. . . a technique that stations many cops in areas with higher crime rates; these areas overlapped with areas inhabited by lower-class minorities. Police initially utilized this technique to prevent crimes from happening in hot spots, but the specific measures that would be taken to prevent crime were often left unclear; there were almost no boundaries to these officers’ powers as authority figures who could stop at nothing in their crime-fighting efforts, which ironically led to many officers committing brutal crimes themselves.
Sarah Childress (2016), in an article for Frontline, argues similarly that broken windows policing has led to aggressive over-policing of non-White communities:
Such practices can strain criminal justice systems, burden impoverished people with fines for minor offenses, and fracture the relationship between police and minorities. It can also lead to tragedy: In New York in 2014, Eric Garner died from a police chokehold after officers approached him for selling loose cigarettes on a street corner. Today, Newark and other cities have been compelled to re-think their approach to policing. But there are few easy solutions, and no quick way to repair years of distrust between police and the communities they serve.
Do specific types of proactive policing strategies lead to lawless behavior of the police? How do proactive policing strategies affect the communities served by the police? Do they lead to higher or lower evaluations of police legitimacy? Do they affect community cohesion more generally? Do they lead to inequitable policing practices that target specific ethnic or racial groups? These are key questions that have not been reviewed systematically across the range of proactive policing strategies. Moreover, these strategies vary widely and thus might be expected to have differential impacts on these outcomes.
This report addresses these questions regarding proactive policing. It reviews what is known about the consequences of proactive policing for crime control, communities, legality, and racial disparities and racial bias. Below we state the specific charge to the committee and then provide a historical review of proactive policing in order to place the report in context. We conclude this introductory chapter with a discussion of the specific definition of proactive policing used by the study committee in framing its report, followed by a summary of the organization of the report.
In 2004, as noted above, the NRC published a report, Fairness and Effectiveness in Policing: The Evidence, which reviewed the existing evidence on police effectiveness and rebutted what had been a longstanding belief that the police had only a limited capacity to prevent crime. However, only a small number of proactive policing strategies were reviewed in that report, and since 2004 a substantial number of studies have assessed the effectiveness of proactive policing strategies. The time is right for a more comprehensive evaluation of proactive policing that includes not only its crime prevention impacts but also its broader implications for justice and U.S. communities.
The National Institute of Justice and the Laura and John Arnold Foundation asked the National Academies of Sciences, Engineering, and Medicine to review the evidence regarding the consequences of different forms of proactive policing for crime and disorder, discriminatory application, legality, and community reaction and receptiveness. The National Academies appointed the Committee on Proactive Policing: Effects on Crime, Communities, and Civil Liberties to carry out this task. Fifteen prominent scholars representing a broad array of disciplines—including criminology, law, psychology, statistics, political science, and economics—as well as two noted police practitioners were included on the committee, which met six times over a 2-year period. The specific charge to the committee was stated by the National Academies as follows:
An ad hoc committee under the auspices of the National Research Council’s (NRC’s) Committee on Law and Justice (CLAJ) will review the evidence on: (1) the effects of different forms of proactive policing on crime; (2) whether they are applied in a discriminatory manner; (3) whether they are being used in a legal fashion; and (4) community reaction. The committee’s review of the literature and the subsequent report will include a thorough discussion of data and methodological gaps in the research.
Attention to proactive policing as a broad-based police organization approach to reduce crime in communities is a relatively recent phenomenon in American policing. The use of the term “proactivity” did not develop until the 1960s, and a focus on the idea that the police would be proactive in efforts to do something about crime was not one to be found often in the policing literature until recent decades.1 Indeed Sam Walker, a noted
historian of the police, concluded that proactive policing “has almost no history prior to the late 1970s” (Walker, 2016, p. 1).
The term “proactive policing” was coined by Albert J. Reiss Jr. and David Bordua as part of a more general examination of the nature of police organization (Bordua and Reiss, 1966; Reiss and Bordua, 1967). They argued that different types of police organization would be needed to deal with different types of police activities. Reactive strategies were seen as those that required simply that the police respond to a citizen request for service. Such activities of the police were seen as more easily managed by a centralized command structure and enjoyed a measure of legitimacy because police were mobilized at the request of a citizen seeking police assistance (Reiss, 1971). However, because police practices that involved proactivity were initiated without a specific request for police involvement, they demanded a more professional and regulated style of police organization, since they involved a wider array of activities involving greater autonomy of police officers.
Of course, police proactivity, as defined by Reiss and Bordua, occurred long before scholars introduced it to the academic and practitioner lexicon. Some police have always been proactive on an individual level, as a matter of personal choice. In this context, many types of activities carried out by police officers throughout the past century have been proactive in that they have used proactive approaches to respond to identified problems. Moreover, as Bordua and Reiss (1966) pointed out in identifying the importance of proactivity, calls from citizens generate a reactive response, but vice offenses infrequently generate complaints, and so vice enforcement requires a degree of proactivity on the part of police officers. For example, when a 19th century foot patrol officer decided on his own to roust public inebriates because they might disrupt commerce on his beat, he was engaging in a proactive type of policing. But when Walker (2016) talks about the virtual absence of proactive policing from the landscape of American policing, he is referring to proactive policing as an organizational crime-prevention strategy, one that began to develop in the latter part of the 20th century, not as a tactic selected independently by a street-level officer or out of an informal culture of policing (see National Research Council, 2004; Weisburd and Braga, 2006a). Proactive policing has come to refer to an expansion of the practices of the police beyond simply responding to and investigating crime; and it takes a strategic approach to crime problems, meaning that these are strategies that were seen as intentional policies of police organizations to develop effective crime control.
Walker (2016) found that the available source material on policing in the 19th century is extremely limited. Nonetheless, the materials that do exist, with some exceptions, indicate an absence of any police organization–directed activities that might be considered even remotely proactive in the
contemporary sense. Scholars are unanimous in characterizing U.S. police organizations in this period as dominated by politics, corrupt, inefficient in terms of crime control, and marked by uncontrolled abusive practices against people on the street (Miller, 1977; Fogelson, 1977; Walker, 1977).
It is important to understand that, despite the nominal quasi-military structure of police departments, U.S. police in the 19th century were in fact extremely disorganized in the sense of modern bureaucracy (Reiss, 1992). With little centralized direction from police chiefs, police commanders simply did not think about proactive efforts to address crime and disorder (and it should be remembered that U.S. cities in this period, with large numbers of recent immigrants and high rates of transiency, were extremely disorderly). To the extent that police agencies in the United States were proactive, they showed initiative in helping to turn out the vote for machine politicians and in discouraging the vote for the opposition (Haller, 1976; Miller, 1975). In fact, much of the police proactivity that actually focused on offenses was intended to promote or protect crime, such as the regulation of thieves and selective law enforcement favoring some over others, all for the financial benefit of the police or the political benefit of their partisan machine allies. Beyond a simplistic belief that patrol deterred crime, there is no evidence of serious thinking about how the police might control crime and disorder more effectively. There was no effort devoted to professional police administration. The idea that the police were public servants, with a broad mission to serve and protect, did not crystalize until the early 20th century with the advent of the police professionalization movement. One manifestation of that development was the first book on police administration, which was published only in 1909 (Fuld, 1909).
Of course, this is not to say that proactivity in policing was absent. It was present in antebellum American policing even before the creation of unified municipal police accountable to a single authority (e.g., mayor) with full-time employees and a structure of internal hierarchical accountability. Levett’s (1975) historical analysis shows that even in times and places where “entrepreneurial” forms (constables, city marshal, high constable, night-watch, day/night police)2 provided diffused modes of policing delivery,3 the proactive control of “disorderly” people4 constituted a significant portion of documented police activity. And following the unification of American
2 Internal organizational hierarchy played a very limited role in regulating activities; officers competed for rewards, and work focused on protecting and recovering property for a fee (Reiss, 1992, p. 69).
4 This included dealing with public drunkenness, prostitution, lewdness, vagrancy, vice, domestic disturbances, doing Sunday business, keeping an untidy house, workingmen strikes, and slavery runaways (Levett, 1975, pp. 52, 114).
municipal agencies, the number of arrests for such offenses, especially drunkenness, rose sharply.5 Levett argues that following the unification of American policing into “political bureaucracies,” local elites used the police to stigmatize and control immigrants and the lower classes, which were perceived to be the source of riotous, immoral, and disorderly behavior. Some might be tempted to draw connections between this and the emergence of “broken windows” as a proactive police management strategy that also focuses on disorders and that emerged in the l980s. But the progenitors of the broken windows approach articulated a detailed logic model justified by crime prevention, not the control or suppression of “dangerous classes” (Wilson and Kelling, 1982).
Another important point is that the will and capacity of police administrators to impose strategies effectively was dependent upon the emergence of a “police civil service bureaucracy,” which only began to emerge in the early 20th century and was characterized by a great reduction in the influence of political elites, replacement by an elaborated police hierarchy, and the codification of personnel policies (civil service) (Reiss, 1992, pp. 70–73). It took many decades for a truly legalistic, technocratic police bureaucracy to take hold in the United States (Reiss, 1992, p. 82) so that the prerequisites for strategic proactivity were feasible.
Other examples of proactivity in early American police departments include the corrupt methods of the police in organizing and regulating thieves and pickpockets. Indeed, public negativity about proactive crime detection by private entrepreneurs motivated the emergence of the modern police detective as an agent who is mobilized only in reaction to the reporting of a crime and who is controlled by the creation of the “case” as a structure of accountability (Klockars, 1985, Ch. 4). Creators of the new police detective in 19th century London were sensitive to the risks to police legitimacy posed by employing the proactive approaches embraced by entrepreneurial private detective agencies, such as the notorious Bow Street Runners. But American police agencies adopted many of these same proactive strategies. They developed networks of criminals as informers, offering immunity from arrest for information on others (Haller, 1976). Thief-taking (for financial reward) produced incentives for taking only cases with good prospects for a large reward, and it encouraged the development of close working relations with professional thieves and fences through whom police shared the rewards with favored criminals. In addition, the practice of “thief-making”
5 American police of the late 19th and early 20th centuries were not only proactive in arresting drunks and the homeless (many of whom were migrants and immigrants), but also proactive in offering them shelter in police stations (Haller, 1976; Monkkonen, 1981). In a limited way, this presaged aspects of community and problem-oriented policing that emerged many decades later.
was publicly unpopular because it employed deceit to entrap or seduce people into engaging in criminal acts. Lincoln Steffens, as well as other muckrakers of the 19th and early 20th centuries, described how American police detectives resorted to these unpopular proactive methods to “license” certain thieves to operate (in exchange for a share of the proceeds of their work) while enforcing the law against others (Steffens, 1931, pp. 222–223). Con artists were required to pay bunco squad members a fee for non-enforcement. Another proactive tactic was to repeatedly harass a thief with a vagabond arrest until he left town (Haller, 1976). And dragnet arrests, made in response to a highly visible crime or crime wave, brought in many people innocent of the crime. While these methods were undeniably proactive, they can hardly be characterized as justified as primarily crime preventive, and they do not constitute a model or positive precursor to the sorts of contemporary proactive innovations the committee has targeted for evaluation.
The police professionalization movement that emerged in the early 20th century had a powerful and long-lasting impact in transforming local police departments and routine policing. The movement had a clearly focused reform agenda that included articulating a clear mission in society, as befits a profession; eliminating the direct political influence that had underpinned the corruption and inefficiency of the police in the 19th century; securing skilled administrators as police chief executives; introducing the principles of modern management to police organizations; and raising personnel standards with regard to recruitment, training, discipline, and retention.
By the end of the 1950s, after 40 to 50 years of reform efforts, most police departments were far more “professional” than they had been in 1900 or 1910 (Reiss, 1992).6 Although significantly deficient by contemporary standards, they were better managed, with at least a nominal commitment to professional standards; better organized; and with rank-and-file officers who, despite many great deficiencies, were far more qualified than their earlier counterparts (Fogelson, 1977; Walker, 1977). Corruption, although still a problem, was no longer as blatant or pervasive as it had been. However, as the turmoil of the 1960s and beyond quickly demonstrated, many problems had not been addressed. The most serious included racial justice and the control of officer discretion, particularly with regard to the use of
6 The committee uses the term “professional” largely in the sense that police reformers of the time used it: a combination of bureaucratic and professional occupation ideals.
deadly force and physical force and with regard to equal justice in stops, arrests, and employment practices.7
The great changes that occurred during the nearly half century of reform, with some notable exceptions, did not include the development of innovative approaches to the control of crime and disorder of the kind that are associated with proactive policing today. One noteworthy exception was the creation of the first police juvenile units, which also led to the employment of the first female police officers in the United States.8 The new juvenile policing units, pioneered in Portland, Oregon, by Lola Baldwin, represented a proactive approach that sought at the outset to reduce juvenile crime with activities that disrupted the forces driving youths down the pathway to delinquency. The approach had a clear problem-oriented focus on juvenile delinquency, on young girls in particular, and in some instances on prostitution (then generally referred to as “White slavery”). It also involved nontraditional police tactics. Policewomen would patrol movie theaters, amusement parks, beaches, pool halls, and other locations or events that attracted young people, to look for juveniles who appeared to be engaging in or about to engage in illegal behavior (Walker, 1977, pp. 84–94). Their mandate was extremely broad. The head of the Detroit policewoman’s unit explained that “a patrol problem may be defined as any situation, arising in a public place, that is potentially harmful to a woman or child” (Hutzel and MacGregor, 1933, p. 11). A few other innovative, proactive reform programs paralleled the new juvenile units.9
Some of the most notable efforts to promote innovative proactive approaches to crime came from the highly visible and influential progressive police leader, August Vollmer. He mobilized his small police force in Berkeley, California, to engage in raids of gambling and opium dens and later did the same during his short tenure as chief in Los Angeles (Oliver, 2017, pp. 169, 373). Ironically, late in his career, Vollmer (1936) wrote a controversial chapter in his book The Police and Modern Society in which he advocated
7 See the findings and recommendations of both the President’s Commission on Law Enforcement and Administration of Justice (1967) and the Kerner Commission (1968); also see Walker (1998, pp. 180–201).
8 Women had been employed as police matrons in the 19th century, but they were primarily jail officials responsible for female inmates.
9 Particularly notable was the Golden Rule policy initiated by Cleveland Police Chief Fred Kohler in 1908. Kohler was deeply disturbed by the high volume of arrests the police made each year, particularly for minor offenses. “I couldn’t see that these wholesale arrests did any good,” he declared. They not only “did not produce good results,” he added, “they did harm.” The Golden Rule involved what experts would recognize as diversion, de-escalation, and mediation. No juveniles would be taken to jail but instead would be taken home to their parents. Officers were directed to use “kindly efforts” to resolve domestic disputes. Finally, individuals who had broken the law because of “unfortunate circumstances” were to be given a reprimand rather than be arrested (see Walker, 1977, pp. 94–98).
a different form of police proactivity to deal with vice. In this book, he rejected the notion that the police should play a central role in dealing with prostitution, gambling, liquor, and narcotics. He argued that police were corrupted by involvement in enforcing laws against these vices and that these were appetites properly left to medical experts who draw upon insights from scientific research (Oliver, 2017, p. 486). Vollmer also devoted a chapter in his book to crime prevention, offering recommendations that presaged key features of the contemporary proactive strategies of community policing and problem-oriented policing. He advocated getting community leaders outside the police involved in crime prevention, drawing on an analysis of the problem (e.g., early childhood intervention), and working in partnership with other agencies. Even much earlier in his career, as chief of police in Berkeley, Vollmer showed a prescient concern for promoting the legitimacy of the police by what we now would call “procedural justice” in the way he himself dealt with offenders and police officers (Sherman, 2017, pp. xi–xii). But counterbalancing Vollmer’s advocacy of a broader police role in some regards, his books and reports also repeated the standard agenda of police professionalization, especially the central mission of police as crime fighters (Vollmer, 1936; Vollmer and Parker, 1937). Two things are particularly worth noting. First, Vollmer’s innovativeness was seasoned by and a part of the larger police professionalization movement. Second, Vollmer’s innovative inclinations were remarkably exceptional (Sherman, 2017) and did not take hold as an active and vital, broadly based reform agenda until they emerged again about a half-century later.
The emergence of the strategies reviewed in this report can be traced to challenges facing the police in the 1960s.10 During the 1970s, criticisms of the police proliferated, as did criticism of the criminal justice system in general (Weisburd and Braga, 2006b; LaFree, 1998). This wave of criticism in part reflected the heightened level of social unrest experienced in the latter years of the 1960s, unrest that included race riots in urban centers and growing opposition to the Vietnam War, particularly among younger Americans. These forms of social unrest often put their young participants, even those from the middle class, as well as racial minorities, in conflict with the police. But the growing sense of a crisis in policing during this period also reflected fears that the criminal justice system was failing to combat crime in the United States effectively. In 1967, a presidential com-
mission reinforced these doubts about the criminal justice system in its report, The Challenge of Crime in a Free Society:
In sum, America’s system of criminal justice is overcrowded and overworked, undermanned, underfinanced, and very often misunderstood. It needs more information and more knowledge. It needs more technical resources. It needs more coordination among its many parts. It needs more public support. It needs the help of community programs and institutions in dealing with offenders and potential offenders. It needs, above all, the willingness to reexamine old ways of doing things, to reform itself, to experiment, to run risks, to dare. It needs vision. (President’s Commission on Law Enforcement and Administration of Justice, 1967, pp. 80–81)
Shortly thereafter, the National Advisory Commission on Civil Disorders (also known as the Kerner Commission) published a report that raised significant questions about the nature of criminal justice and the organization of policing in the United States. However, the central issue for policing raised in this report was the relationship between the police and racial and ethnic minorities in predominantly non-White communities. Although the report did not focus primarily on the police as responsible for patterns of discrimination against Blacks, it did present the police—as well as other criminal justice agencies—as contributing to those patterns, rather than helping to find solutions to the difficult social issues involved: “In Newark, Detroit, Watts and Harlem, in practically every city that has experienced racial disruption since the summer of 1964, abrasive relationships between police and Negroes and other minority groups have been a major source of grievance, tension, and ultimately disorder” (National Advisory Commission on Civil Disorders, 1968, p. 157).
In response to both the concerns documented in these two reports and the growing sense of alienation between the police and the public in the latter half of the 1960s, policy makers, the police, and scholars increasingly questioned the adequacy of how American policing was organized, particularly with respect to the strategies that had dominated American approaches to policing since at least World War II. The NRC has characterized these approaches as the “standard model” of policing:
This model relies generally on a “one size fits all” application of reactive strategies to suppress crime, in contrast to more customized and proactive strategies. The standard model also emphasizes the role of arrests and the threat of punishment in achieving this objective, with less emphasis on other capabilities of the police. The standard model of policing has assumed that generic strategies for crime prevention can be applied throughout a jurisdiction, regardless of the level of crime, the nature of crime, or other possible variations. (National Research Council, 2004, p. 223)
General types of strategies that have been prominent in the standard model of policing include increasing the size of police agencies, random patrol across all parts of the community, rapid response to calls for service, generally applied follow-up investigations, and generally applied intensive enforcement and arrest policies (National Research Council, 2004, p. 224).
The standard model of policing was primarily a reactive model. Its focus on follow-up enforcement, rapid responses to citizen calls to the police, and investigation of crime and apprehension of criminals are directly responsive to the commission of a crime or citizen notification of crimes occurring. Even random preventive patrol, which was seen as a key method for deterrence of crime through the visible presence of police across a city (Repetto, 1976; Kelling et al., 1974), was rooted in the necessities of the rapid response system. With the advent of radio dispatch responses to emergency calls to the police, a key factor was having police cars spread in a jurisdiction to allow the police to respond to calls quickly. Accordingly, the standard model of policing was strongly rooted in the police reaction to a crime being committed.
Although important issues were being raised about the standard model of policing well before the end of the 1960s decade, there was at that time a relative dearth of academic research on the impacts of the policing strategies then in vogue on crime rates or on how the public viewed the police. The prevailing attitude was that post–World War II policing practices incorporated major improvements over policing strategies of prior decades and that these practices were effective not only in responding to specific crime events but also in having overall impacts on crime in the jurisdictions that police served. The crime control benefits were seen as resulting from the deterrence gained by police effectively identifying and investigating offenders, responding quickly to the scene of crimes, and being visible agents of control as they organized themselves for the new rapid response systems that radios and police cars enabled. But the issues identified during the 1960s showed the need for research on the standard model, and serious attention to that research began in the 1970s.
Since the founding of the London Metropolitan Police in 1829, modern policing had been grounded in Sir Robert Peel’s principle that the police could effectively control crime through visible patrol dispersed through the larger community and organized by assigning officers to specific police beats and holding them accountable for patrolling those beats (Grant, 2010; Critchley, 1972). The assumption was that a visible police presence would deter criminals from offending. Additionally, dispersal of patrol throughout the community would make officers readily available to respond to problems they observed or were asked to deal with.
A large Police Foundation study in the 1970s sought to establish whether evidence actually supported these broadly accepted assumptions
regarding visible police patrol. The study was one of the first large field trials in American policing. Although the design of this study was subsequently criticized (Larson and Cahn, 1985; Minneapolis Medical Research Foundation, Inc., 1976; Sherman and Weisburd, 1995), its results were to have lasting impact on assumptions regarding the impacts of policing on crime. Conducted in Kansas City, Missouri, preventive patrol was manipulated in large beat areas, with areas having higher, lower, or standard levels of police patrol vehicles. The study concluded that merely increasing or decreasing the intensity of routine preventive patrol by police officers in cars had no effect on crime, on delivery of police services to citizens, or on how community members felt about security in their communities.
Another large-scale study, conducted by Spelman and Brown (1984), challenged a core assumption of the standard model of policing: namely, that a more rapid response to calls for service would improve crime outcomes. A prior investigation in Kansas City had found little support for the crime-control effectiveness of responding more rapidly to calls for service (Kansas City Police Department, 1977), and the Spelman and Brown study was designed to test that assumption with greater rigor. With support from the National Institute of Justice, the research team interviewed 4,000 individuals who had been victims, witnesses, or bystanders in about 3,300 serious crimes committed in four U.S. cities. Based on the data they collected, these researchers challenged the crime-control effectiveness of rapid response to calls for service:
Rapid police response may be unnecessary for three out of every four serious crimes reported to police. The traditional practice of immediate response to all reports of serious crimes currently leads to on-scene arrests in only 29 of every 1,000 cases. By implementing innovative programs, police may be able to increase this response-related arrest rate to 50 or even 60 per 1000, but there is little hope that further increases can be generated. (Spelman and Brown, 1984, p. xix)
Another element of the standard model, the use of follow-up investigations by police, was examined in a series of empirical studies in the 1970s and early 1980s. An assumption of the standard model was that general improvements in the methods used in police investigations would help to control crime for two reasons: more of the active offenders would be in prison, where they would no longer be committing crimes in the community; and the prospect of being discovered and arrested would deter potential offenders (National Research Council, 2004). However, the empirical studies during this period found that follow-up investigations had little effect on crime rates (Eck, 1983; Greenwood et al., 1975; Greenwood, Petersilia, and Chaiken, 1977; Skogan and Antunes, 1979).
In understanding the emergence of proactive policing, it is important
to recognize the impact that these studies had on scholars and police at the time. In retrospect, however, many scholars overstated what could be learned from the findings about standard police practices (see, e.g., Goldstein, 1979; Gottfredson and Hirschi, 1990; Bayley, 1994). And some evaluations during this period reported more positive results from such standard police practices as routine preventive patrol (see, e.g., Chaiken, 1978; Press, 1971; Schnelle et al., 1977). Moreover, the body of research on the standard model of policing that has developed since these early studies provides a more nuanced portrait of that model’s crime prevention outcomes.
This is especially the case in considering whether police staffing levels influence levels of crime. Econometric studies that make strong efforts to overcome key measurement and specification problems have begun to show significant crime-prevention gains for increases in the number of police in a city (see, e.g., Evans and Owens, 2007; Machin and Olivier, 2011). However, the conclusion that these studies reflect the impact of the standard model of policing has been criticized because they often examine the boost in police resources that comes from support for community policing or other proactive policing strategies (Lee, Eck, and Corsaro, 2016). At the same time, studies of police strikes conducted in periods when the standard reactive model of policing was dominant suggest that crime does go up in the absence of police (Sherman and Eck, 2002; Nagin and Weisburd, 2013). While the committee recognizes the importance of these studies as well as the more general questions raised regarding the impacts of the standard model of policing on crime, we do not draw a conclusion about its crime prevention outcomes. However, given the continued importance and dominance of the standard model of policing, we do think that this is an important area for future study.
As the United States entered the 1990s, there appeared to be a scholarly consensus that traditional reactive police practices did not work in preventing or controlling crime (Weisburd and Braga, 2006b, p. 9). For example, Gottfredson and Hirschi stated in A General Theory of Crime, “No evidence exists that augmentation of patrol forces or equipment, differential patrol strategies, or differential intensities of surveillance have an effect on crime rates” (Gottfredson and Hirschi, 1990, p. 270). And a few years later, David Bayley made an even stronger assertion:
The police do not prevent crime. This is one of the best kept secrets of modern life. Experts know it, the police know it, but the public does not know it. Yet the police pretend that they are society’s best defense against
crime . . . this is a myth. First, repeated analysis has consistently failed to find any connection between the number of police officers and crime rates. Secondly, the primary strategies adopted by modern police have been shown to have little or no effect on crime. (Bayley, 1994, p. 3)
Official crime statistics, widely available to the public, seemed to reinforce this view of the ineffectiveness of policing strategies, as well as the general perception that the police were losing the “War on Crime.” Even the established, professional police organizations in America’s largest cities seemed unable to curtail the alarming rise in crime rates—especially violent crime rates, which doubled between 1973 and 1990 (Weisburd and Braga, 2006b, p. 10).
Proactive policing grew out of this period of crisis for American policing. Proactive policing was a product—one of many products, in fact—of an extraordinary convergence of several legal, social, and political crises that swept over American society in the tumultuous 1960s, profoundly affecting the police along with every other institution. The crises generated new demands on the police to improve both their capacity to address crime and disorder and their own internal standards of accountability. The crises of the 1960s were followed, as noted above, by several major research findings that undermined the basic principles that had guided modern policing since the founding of the London Metropolitan Police by Robert Peel in 1829. The result was a period of intellectual ferment as police chiefs, outside experts, and academics searched for new principles for police operations. This search generated numerous innovative responses, responses that came to be termed “proactive policing” and that are reviewed in this report.
The committee believes its task must be seen in historical context and that its definition of proactive policing should be geared to innovations in police practices and policies that have been developed over the past few decades. In this report, the term “proactive policing” is used to refer to all policing strategies that have as one of their goals the prevention or reduction of crime and disorder and that are not reactive in terms of focusing primarily on uncovering ongoing crime or on investigating or responding to crimes once they have occurred. Specifically, the elements of proactivity include an emphasis on prevention, mobilizing resources based on police initiative, and targeting the broader underlying forces at work that may be driving crime and disorder. This contrasts with reactive policing, which involves an emphasis on reacting to particular crime events after they have occurred, mobilizing resources based on requests coming from outside the police organization, and focusing on the particulars of a given criminal
incident. In practice, policing strategies range along a continuum between pure proactivity and pure reactivity. The more proactive elements that are present in a given strategy, the more proactive it is. The more reactive elements present in a given strategy, the more reactive it is.
The committee recognized at the outset that there is no accepted definition of proactive policing among scholars or the public. In the earliest references to proactivity (see Bordua and Reiss, 1966; Reiss and Bordua, 1967), scholars were focused primarily not on the strategies that were subsumed by the definition but rather on the implications of proactivity for the legitimacy of police intrusion in the lives of citizens (Black, 1971; Reiss, 1973). Proactivity was simply the situation where police powers were mobilized not as a result of citizen requests to the police but rather due to the decision, usually by street-level police officers or special units, to initiate enforcement or other policing powers. Proactive mobilization of police resources, as contrasted with reactive mobilization, was seen as creating additional challenges to the public acceptance of police powers because it meant that the police did not have the assent of the public before taking action.
Our definition of proactive policing is consistent with earlier conceptualizations of this idea in that we focus on situations where the mobilization of police resources comes as a result of the initiative of the police and not of citizens. Accordingly, proactive policing as we define it raises many of the questions about mobilization of police resources without citizen requests that interested these early policing scholars. However, proactive policing, in contrast to proactivity itself, refers to a group of strategies and programs, many of them initiated over the past three decades, for preventing crime.
As we noted above, the 2004 NRC report on police practices and policies proposed what it termed the “standard model of policing” to describe the common ways in which policing was organized before the 1980s. The study committee for that report drew from Herman Goldstein’s classic critique of American policing in his article on problem-oriented policing published in 1979 (see also Goldstein, 1990). In that article, he tried to understand why a series of studies of American policing in the previous decades seemed to show that policing was ineffective in preventing crime. His conclusion was that policing had begun to focus more on the means of policing than its ends. Policing in this context had become focused on how fast the police could respond to calls for service, not how it could structure its responses to be most effective in reducing crime. Police managers had become concerned primarily with how to get enough officers on the street to meet their geographic patrol obligations and not upon how the allocation of patrol could be used most effectively to prevent crime.
The NRC study committee (National Research Council, 2004) identified two main ways in which innovative proactive strategies moved beyond the standard model of policing. The first is that many of the new strategies
used “focus” in efforts to prevent crime. Many strategies identified specific geographic areas, for example crime hot spots, that would receive greater police attention. Other strategies capitalized on the fact that high-rate offenders were responsible for a large proportion of the crime problem. Accordingly, one key factor that distinguished innovative policing strategies was their approach to identifying how to focus resources on particular places and people. That study committee viewed this as one component of proactivity. In this case, the police do not simply comply with their reactive obligations to respond to and investigate crimes; rather, they purposely and strategically focus such resources to prevent crime.
The new proactive strategies went beyond the obligations of the police to respond to the occurrence of crime and to investigate and bring offenders to justice and focused instead on policing practices that could be successful in preventing crime irrespective of whether they had been seen in the past as traditional components of police practice. Because of this, the 2004 NRC report also identified an expansion of the tools of policing as an important innovation in police practices over the standard model (National Research Council, 2004, pp. 84–93, 232–251). The new proactive policing strategies pioneered a wide variety of new tools, ranging from community collaborations to the use of civil ordinances and to the introduction of innovative technologies that bring new information to enhance crime prevention.
But the new proactive policing strategies also reinterpreted traditional practices of policing to advance the crime control mission. For example, general preventive patrol is a key element of the standard model of policing. Innovative proactive policing strategies drew upon patrol methods but changed their mission through the development of hot spots policing (Sherman and Weisburd, 1995). In this case, police patrol in motorized vehicles, a key component of policing since the 1940s, was reallocated to specific places where crime was concentrated, in a conscious effort to be more successful in preventing crime. Stopping and questioning citizens had been a part of the standard practices of policing long before the Supreme Court specifically allowed it as a policing approach in Terry v. Ohio in 1968. However, this committee’s interest in the practice called “stop, question, and frisk” develops not from the practice itself but rather from its use in some jurisdictions as a strategic proactive approach for anticipating and preventing crime.
There are likely scores of innovative proactive policing approaches that have been tried in police agencies in the United States and abroad. The committee could not review them all in depth, so we accordingly made a decision to give priority to certain types of proactive policing strategies. The first type includes strategies that have become commonly applied in American police agencies. It seemed important to us to provide insight into the effectiveness and potential intended and unintended impacts of proac-
tive strategies that are already widely adopted in American police agencies. At the same time, we wanted to assess new and innovative practices and policies that may not yet have been widely adopted but seemed to the committee to represent important potential strategies for policing efforts to prevent crime. Finally, policing is in a period of tremendous community concern. Some of that concern is focused on proactive policing strategies that are seen as unfairly targeting some Americans over others and as leading to abusive policing practices. Accordingly, in selecting the specific practices and programs that would be examined by the committee, we agreed to focus particular attention on those that had been criticized for leading to biased or abusive outcomes or that sought to use positive community engagement as a method of enhancing crime control.
The committee decided not to examine innovations that were primarily technical in nature and did not include a clearly articulated goal of preventing crime. Some of these innovations—for example, computerized crime mapping—are often strongly linked to proactive policing innovations. These are included in our review in the context of those innovative strategies. But other new technologies being adopted by the police, such as body cameras or drones, do not as of yet have a specific strategic connection to crime control or proactive policing. We agree that such approaches should be assessed and reviewed (see, e.g., Lum, Koper, and Willis, 2016), but such a review goes beyond the scope of this report.
The committee included scholars from different disciplines, which sometimes emphasize different methodological and analytic approaches to developing evidence. Because of this, the committee took a broad approach in applying standards of evidence and included within its purview, for example, experimental studies, rigorous quasi-experimental approaches, econometric methods, and legal analysis. However, the committee also was in overall agreement regarding the characteristics of studies that would make the evidence persuasive for drawing conclusions.
A number of templates have been suggested for making systematic judgments about the strength of the statistical evidence in the case of a single evaluation or study, as, for example, the template incorporated in the What Works Clearinghouse established by the U.S. Department of Education’s Institute for Education Sciences. Closely related are the templates for addressing the strength of evidence from a series of studies on the same sort of intervention, as with the Campbell Collaboration systematic reviews in education, crime control, parenting, and other areas. These reviews have been conducted for some categories of proactive policing (e.g., hot spots policing, problem-oriented policing, and focused deterrence policing), and
they have informed the work of the committee. However, we have chosen not to rely upon a formal process of this sort in preparing this report. Instead, our approach focused on the committee reviewing the available evidence in each area and then providing an in-depth critique of studies’ methods and conclusions. Here, we sketch the main considerations relevant to assessing the strength of evidence, considerations that guide both the committee’s critiques and the statistical evaluation templates used elsewhere.
In considering the evidence from a single field test of an intervention, there are two main tasks. The first is to determine how informative the study is regarding the causal impact of the intervention on designated outcome variables in the current field test. The second is to determine the extent to which the results from this particular field test can be extrapolated to policing more generally. In the usual parlance, the first task concerns the internal validity of the impact evaluation, whereas the second task assesses its external validity. The statistical science associated with judging internal validity is well developed and is often easier for the committee to assess in our review. Yet the external validity of a finding or set of findings is particularly important in policy analysis, where the goal is to use the research evidence to shape policy development. In our review, we considered in a general way whether we can draw more general inferences about policing from specific studies. In some cases, that led us to note the limitations of, for example, using laboratory studies to make claims about police behavior in the field. In other cases, such as for hot spots policing studies, we note the large number of studies conducted in different contexts. A large group of experiments conducted in different places, in different types of police agencies, for example, provides a more convincing argument for the external validity of study findings than one or a small group of studies that have been conducted in one city. The limitations in the research base in policing means that we have to be cautious in drawing specific policy recommendations for police agencies. We return to this important issue in our detailed discussion of policy implications in Chapter 8.
The first task noted above, developing an internally valid estimate of the causal impact in a particular field test, requires outcome data of acceptable quality; both random and systematic errors in measurement are of concern. Next, a valid estimate of what levels those outcomes would have taken if the intervention were not implemented is required. These alternative values are called potential outcomes or counterfactuals. The “effects” of interest are defined as the difference between the observed values and the counterfactual values.
There are a variety of methods (“study designs”) available for estimating the counterfactual values. In general, randomized controlled trials (RCTs) are seen as providing the strongest approach for creating such
estimates. In an RCT, some units of observation are randomly assigned to the intervention and others are assigned to a control group receiving the alternative to which the treatment is being compared. The outcomes of the control group are then used to estimate the counterfactual for the treatment group. In principle, this approach ensures that the assignment of the treatment is not correlated with the potential outcome (which would impart bias to the estimated impact). A well-done RCT with reliable outcome data provides an unbiased estimate of the causal effect of the treatment, together with an estimate of how much statistical uncertainty is associated with that effect estimate.
In practice, an RCT may be difficult or even unethical to implement in a particular setting, or this design may engender administrative–fidelity problems that cloud the validity of the estimate of effect (e.g., cross-over from one condition to the other, noncompliance with treatment assignment, or treatment spill-over). There are alternative “quasi-experimental” research designs that in some cases may also produce trustworthy estimates and, indeed, share key statistical properties with high-quality RCTs (Nagin and Weisburd, 2013). These designs, when rigorous, identify methods for developing plausibly “as good as random” comparisons to use as the counterfactuals to the treatment condition. Natural experiments are examples of such research designs (Cook and Campbell, 1979), as are regression discontinuity designs (see, e.g., Berk, 2010). As another alternative to an RCT design, there are studies that use statistical controls as a primary method for providing valid estimates of the impacts of interventions. These are often termed multivariate methods, but they may mimic other types of quasi-experimental designs (e.g., propensity score matching, described by Rosenbaum and Rubin, 1983). These studies rely on high-quality data about the phenomenon under study, as statistical models are used to create equivalence of treatment and control conditions by including alternative confounding explanations of observed differences between treatment and nontreatment outcomes as statistical controls. Thus, assessing internal validity for all of these approaches requires a close understanding of the data-generating process. While the committee recognized the inherent advantages of randomized experiments, it assessed the strengths of specific studies in terms of how well threats to their internal validity had been addressed.
The second task (external validity) involves determining how relevant a particular finding or set of findings regarding an intervention’s effectiveness is to estimating the potential effectiveness of similar interventions in other times and places. The challenge for this task is that while the new interventions are “similar” in some sense to those that were evaluated, they and the context in which they are implemented will not be identical to the evaluated cases. For example, if a hot spots policing intervention is effective
in reducing robbery in a high-crime neighborhood in Chicago, would it also be similarly effective in reducing robbery in a high-crime neighborhood in Los Angeles? The implementing agency and the environment both differ in a variety of ways between these two neighborhoods. Does that negate the relevance of the Chicago finding? More generally, changes in the treatment details, the way in which it is implemented, the context of the implementation, and differences in the populations exposed can have considerable effects on the impact. Despite these potential pitfalls, for the purposes of policy design it is necessary to extrapolate from one time and place to a different time and place.
One way to strengthen the credibility of extrapolation is to show that the findings in that Chicago neighborhood can be replicated through high-quality evaluations in a number of other cities. That is, if the finding seems robust with respect to some other times and places, then it is more credible to extrapolate to still others. Alternatively, the intervention effect may vary, but in systematic ways. For example, if there is a reasonable presumption that certain factors (such as size and average education of the jurisdiction population) moderate the magnitude of the intervention effect, then the ideal evidence base would include high-quality evaluations conducted in a number of jurisdictions that differ with respect to those moderating factors. In principle that would provide a “predicted effect size” for any jurisdiction of particular size and education.
Another way to strengthen the credibility of extrapolation is by development of theory regarding the basic mechanisms on which a program innovation relies to influence behavior. What is learned from empirical studies of one or more interventions can then be framed as evidence not merely about the effectiveness of the specific interventions but rather about the effectiveness of the mechanisms underlying those interventions (Ludwig, Kling, and Mullainathan, 2011). That is, a series of empirical evaluations, perhaps taken together with other sorts of evidence, can allow evaluators to look inside the “black box” of a policing approach (e.g., hot spots policing) and interpret observed results in terms of the underlying mechanism (e.g., deterrence via the threat of punishment communicated by police presence). The accumulation of evidence supporting the strength and robustness of a particular mechanism enhances confidence that programs in new times and places that incorporate this mechanism will be effective.
These evidence-accumulation strategies rely on the intervention having homogeneous effects that are in fact not context dependent. If multiple studies result in conflicting evidence on effectiveness, new empirical work focusing on uncovering and testing contextual factors that aid or hinder treatment effectiveness is needed.
This chapter has discussed the historical context of proactive policing, the charge to the study committee, the definition of proactive policing used in this report, and the standards used by the committee in evaluating evidence. Chapter 2 focuses more directly on the nature of the proactive policing strategies examined in the report. These strategies are divided into four broad categories: place based, person focused, problem solving, and community oriented. The logic for this division is presented in that chapter, as are the descriptions of the strategies that fall under each of those domains. As will become apparent, the real world is much messier than an academic effort to define and categorize proactive policing strategies. Nonetheless, the committee thought it important at the outset to try to identify strategies in terms of the broad mechanisms that are seen as contributing to crime prevention outcomes.
Policing strategies raise important issues regarding legality and lawfulness. Proactive approaches can involve, among other things, the gathering and aggregating of information, the use of algorithms (public and private) for decision making, the development of criteria for intervention beyond individual suspicion, and the concentration of interventions and resources. Such activities may create concerns about issues, such as privacy, arbitrariness or abuse (including arbitrariness or abuse with regard to arrests and the use of force), discrimination, accuracy, accountability, and transparency. These issues are the focus of discussion in Chapter 3.
The importance of reviewing the evidence of the effects of proactive policing on crime and disorder has already been noted. More than a decade has passed since the 2004 NRC report on police practices and policies (National Research Council, 2004), and many innovations in proactive policing had not been evaluated at the time of that study; other approaches, moreover, have yielded many new studies. These topics are the focus of Chapter 4.
As indicated above, even if the evidence were clear that proactive policing strategies are effective at reducing crime and disorder, the consequences of such strategies would need to be evaluated along additional dimensions. Police officers are some of the most visible representatives of law and government in most people’s lives, and the fairness of policing has become a key issue today. The President’s Task Force on 21st Century Policing, which was established by President Obama in December 2014, emphasized, “[b]uilding trust and nurturing legitimacy on both sides of the police/citizen divide is the foundational principle underlying the nature of relations between law enforcement agencies and the communities they serve” (p. 1). Proactive policing strategies can increase the points of contact and interaction between police and communities, and proactive approaches
may also expand the police function beyond traditional law enforcement activities. The implications of proactive policing policy for community trust and legitimacy are therefore especially important.
Chapters 5 and 6 focus on the impacts of proactive policing strategies on communities and on community perceptions of the police. “Community outcomes” is a term used in this report to refer to how a group of people perceives and feels about its police, the policing that it receives, and the consequences of that policing. It also includes actions that community members take to assist police or to benefit themselves directly to deal with crime, disorder, and quality-of-life issues relevant to policing. We divide this discussion into two chapters to reflect the important distinction between strategies that are focused on crime control without a clear orientation to the community and its role in policing and those strategies seek to use community engagement to enhance crime control. Chapter 5 examines how proactive policing strategies that focus on places, people, or problem-solving impact the communities in which they are carried out. Chapter 6 examines proactive policing strategies, such as community policing and procedural justice policing, that seek not only to reduce crime but also to alter the fundamental relationships between the police and the communities they serve. Clearly, these proactive policing programs would be expected to have more direct, and at least in their logic model, more positive impacts on community perceptions of the police.
Concerns about racial discrimination loom especially large in discussions of policing. There are many historical reasons why non-Whites might distrust law enforcement. For instance, when the laws of the United States were designed to produce and maintain racial stratification, it was the job of police officers and sheriff’s deputies to enforce those laws. Police across the nation were tasked with enforcing laws that disadvantaged Blacks, Native Americans, immigrants, and others who were targeted by laws designed to reinforce notions of racial superiority. From the Fugitive Slave Act of 1850, which regulated the movement of Black people before emancipation, to sundown towns that required all non-Whites to leave a jurisdiction before the sun set, and to segregated schools, water fountains, and lunch counters, it was the job of law enforcement to regulate de jure racial hierarchies (Hinton, 2016a, 2016b).
Beginning in the 1960s and 1970s, law enforcement tactics under the “War on Crime” and “War on Drugs” were characterized, if not by racial animus, then by racially disparate consequences (Hinton, 2016a, 2016b). More generally, even scholars trying to reform the police often seemed to neglect the question of race and the impacts of policing on non-White communities (Williams and Murphy, 1990). And this concern with discrimination and disparate consequences for non-White communities has continued through the new millennium. We review in Chapter 7 not only the evidence
on explicit biases against Black and other non-White people but also implicit biases that may play a role in policing even when the police have no specific policies to target non-White individuals.
A parallel (if less prominent) critique of police and race in the United States is that Black neighborhoods suffer from under-policing. Anything that reduced crime—especially violent crime—in non-White neighborhoods would be a boon to those communities. This was precisely the argument advanced by the Clinton administration in support of the 1994 Omnibus Crime Bill that poured federal resources into municipal policing (Brickey, 1995; Hinton, 2016a, 2016b).
This report does not answer a series of questions at the heart of tension between non-Whites and the police across the United States: Is policing biased against the poor, Blacks, or other non-Whites? Are they more likely to be shot and killed than advantaged groups? These are key questions that need to be answered. The focus of this report, however, is more modest. The committee’s main interest is whether and to what extent proactive policing affects racial disparities in police–citizen encounters and racial bias in policy behavior.
Chapter 8 summarizes the main findings for each of the four areas on which the report focuses: law and legality, crime control, community impacts, and racial disparities and racial bias. It then explores the broader policy implications of the report. Finally, it lays out the committee’s suggestions for filling research gaps in order to strengthen the knowledge base regarding proactive policing and its impacts.
During the course of this study, the committee also gathered information through roundtables and webinars open to the public. The purpose of these activities was to explore topics and issues relevant to the study charge from the perspectives of both the police carrying out proactive policing and the communities that experience proactive policing. These sessions, which helped to inform the committee’s deliberations, are summarized in Appendix A.
Proactive policing, as the committee defines it, is a relatively new phenomenon in American cities. Although there were historical precedents for police proactivity in 19th and 20th century America, its current form developed from a crisis in confidence in policing that emerged because of social unrest, rising crime rates, and growing skepticism regarding the standard model of policing that had been dominant in the latter half of the 20th century. The chapters that follow answer the specific questions with which the committee has been charged: What are the consequences of proactive policing for legality, crime, communities, and racial disparities and racial bias?