Tijdschrift voor de politie

These strategies represented fundamental changes to the business of policing. However, as many police scholars and executives point out, improving police performance through innovation is often not straightforward. Police departments are highly resistant to change and police officers often experience difficulty in implementing new programs (Sparrow, Moore, and Kennedy, 1990; Capowich and Roehl, 1994; Sadd and Grinc, 1994). The available evidence on key dimensions of police performance associated with these eight innovations, such as crime control effectiveness and community satisfaction with services provided, is also surprisingly limited.

These observations are not unique to the policing field. For example, as Elmore (1997) suggests, the field of education was awash in innovation during the 1990s, but there is little evidence examining whether those innovations advanced the performance of schools, students, or graduates.
While our knowledge about the effects of these innovations on police performance is still developing, we think there is much reason for optimism about the future of policing. This period of innovation has demonstrated that police can prevent crime and can improve their relationships with the communities they serve. In the near future, we don’t anticipate the dramatic strategic innovations that characterized the last two decades. Rather, we expect further refinement of our knowledge of ‘what works’ in policing, under what circumstances particular strategies may work, and why these strategies are effective in improving police performance. The challenge for the future of policing is to continue making progress in further developing and implementing promising strategies while addressing the new problems of public safety that have been created by 9/11 and the concerns that it has raised about the threat of terrorism and the need for police commitment to homeland security.
 

The Form and Character of Recent Police Innovations
Community policing was one of the first new approaches to policing to emerge in this modern period of police innovation. Community policing programs were already being implemented and advocated in the 1980s (Trojanowicz, 1982, 1989; Kelling and Moore, 1988; Greene and Mastrofski, 1988), and by the 1990s, the idea of community policing had affected most American police agencies. Police practices associated with community policing have been diverse and have often changed over time. Foot patrol, for example, was considered an important element of community policing in the 1980s, but has not been a core component of more recent community policing programs. Community policing has often been implemented in combination with other programs, such as problem-oriented policing, thus making it difficult to distinguish the core components of community policing from those of other innovations that developed during this time period.
One core element of the community policing movement was that the community should play a central role in defining the problems the police address, and that these problems should extend much beyond conventional law enforcement. As Kelling and Moore (1988: 4) argue, ‘during the 1950s and 1960s, police thought they were law enforcement agencies primarily fighting crime.’ In the ‘community policing era’, the police function broadens and includes order maintenance, conflict resolution, provision of services through problem solving, as well as other activities’ (Kelling and Moore, 1988: 2). One way to understand the early development of community policing is to recognize that it responds to the question: What is the justification for the police if they cannot prevent crime? While crime fighting has increasingly become a central concern in community policing over the last decade, an important contribution of community policing to police innovation was its recognition that there were many critical community problems that the police could address that were not traditionally defined as crime problems. The expansion of the police function was to become an important part of many of the innovations discussed in this paper. The definition of new tasks can be seen in part as a response to the failure of police to achieve the crime control goals of the professional model of policing (Kelling et al., 1974; Spelman and Brown, 1984; Greenwood et al., 1977).
Other innovations in policing in this period also looked to redefine the role of the police in one way or another. Broken windows policing also seeks to direct the police to problems that had often been ignored in standard police practices. Wilson and Kelling (1982) identified a link between social disorder and crime which suggested the importance of police paying attention to many problems that were seen in earlier decades as peripheral to the police function. Wilson and Kelling argued that concern with disorder was an essential ingredient for doing something about crime problems. The ‘broken windows’ thesis suggested that serious crime developed because the police and citizens did not work together to prevent urban decay and social disorder. In the context of crime, Wilson and Kelling (1982) argued that ‘untended behavior leads to the breakdown of community controls’ (31). Broken windows encourages the police to be concerned with problems of disorder, and moves crime itself to a secondary, or at least second-stage goal of the police.
Problem-oriented policing also sought to broaden the problems that police-approached. In Herman Goldstein’s original formulation of problem-oriented policing in 1979 he argued that the ‘police job requires that they deal with a wide range of behavioral problems that arise in the community’ (1979: 242). Goldstein suggested that the police could impact crime and other problems if they took a different approach, in this case, the problem-oriented policing approach. In order for the police to be more efficient and effective, they must gather information about incidents and design an appropriate response based on the nature of the underlying conditions that cause the problem(s) (Goldstein, 1990). As summarized by Eck and Spelman: Underlying conditions create problems. These conditions might include the characteristics of the people involved (offenders, potential victims, and others), the social setting in which these people interact, the physical environment, and the way the public deals with these conditions. A problem created by these conditions may generate one or more incidents. These incidents, while stemming from a common source, may appear to be different. For example, social and physical conditions in a deteriorated apartment complex may generate burglaries, acts of vandalism, intimidation of pedestrians by rowdy teenagers, and other incidents. These incidents, some of which come to police attention, are symptoms of the problem. The incidents will continue as long as the problem that creates them persists. (1987: xvi)
And in Goldstein’s words, the problem-solving process requires: Identifying these problems in more precise terms, researching each problem, documenting the nature of the current police response, assessing its adequacy and the adequacy of existing authority and resources, engaging in a broad exploration of alternatives to present responses, weighing the merits of these alternatives, and choosing among them. (1979: 236)
Pulling levers policing strategies adopt a problem-oriented approach, but provides a broader and more comprehensive combination of strategies than more traditional problem-oriented policing programs. Pioneered in Boston to deal with an ‘epidemic’ of youth violence (Kennedy, Piehl, and Braga, 1996), the pulling levers approach, in its simplest form, consists of selecting a particular crime problem, such as youth homicide; convening an interagency working group of law enforcement practitioners; conducting research to identify key offenders, groups, and behavior patterns; framing a response to offenders and groups of offenders that uses a varied menu of sanctions (‘pulling levers’) to stop them from continuing their violent behavior; focusing social services and community resources on targeted offenders and groups to match law enforcement prevention efforts; and directly and repeatedly communicating with offenders to make them understand why they are receiving this special attention (Kennedy, 1997, 2006).
Third party policing offers another solution to the failures of the standard policing model. It follows suggestions made by Herman Goldstein (1979) that the ‘tool box’ of police strategies be expanded. In this case however, the resources of the police are expanded to ‘third parties’ that are believed to offer significant new resources for doing something about crime and disorder. Third party policing asserts that the police cannot successfully deal with many problems on their own, and thus that the failures of traditional policing models may be found in the limits of police powers. Using civil ordinances and civil courts, or the resources of private agencies, third party policing recognizes that much social control is exercised by institutions other than the police and that crime can be managed through agencies other than the criminal law.
Hot spots policing was first examined in the Minneapolis Hot Spots Experiment (Sherman and Weisburd, 1995). Drawing upon empirical evidence that crime was clustered in discrete hot spots (Pierce et al., 1988; Sherman et al., 1989), Sherman and Weisburd argued that preventive patrol might be more effective if it was more tightly focused. If ‘only 3 percent of the addresses in a city produce more than half of all the requests for police response, if no police are dispatched to 40 percent of the addresses and intersections in a city over one year, and, if among the 60 percent with any requests the majority register only one request per year, then concentrating police in a few locations makes more sense than spreading them evenly through a beat’ (Sherman and Weisburd, 1995: 629). Hot spots policing does not demand that the police change their strategies, but requires that they focus them more carefully at places where crime is clustered.
Compstat also responds to the failures of the traditional model by critiquing the ways in which the police carry out their task. However, in the case of Compstat the focus is less on the specific strategies that the police are involved in and more on the nature of police organization itself. If as Herman Goldstein noted in 1979 that the failures of the standard model of policing could be explained by the fact that police organizations were poorly organized to do something about crime, Compstat sought to overcome that pathology. It sought to empower the command structure to do something about crime problems. William Bratton, the New York City police chief who coined the term and developed the program writes:
We created a system in which the police commissioner, with his executive core, first empowers and then interrogates the precinct commander, forcing him or her to come up with a plan to attack crime. But it should not stop there. At the next level done, it should be the precinct commander, taking the same role as the commissioner, empowering and interrogating the platoon commander. Then, at the third level, the platoon commander should be asking his sergeants… all the way down until everyone in the entire organization is empowered and motivated, active and assessed and successful. It works in all organizations, whether it’s 38,000 cops or Mayberry, R.F.D. (Bratton, 1998:239)
Evidence based policing also traces the failures of traditional policing practices to the ways in which the police carry out their tasks. The approach is drawn from a much wider set of public policy concerns, and a broader policy movement concerning the use of rigorous evidence in forming practice (e.g. ‘evidence-based’ medicine). Much police practice is based on tradition and clinical experience, and this is often the only guidance for criminal justice practitioners.
Evidence based policing argues that it is understandable that standard models of policing had failed because successful strategies must be based on scientific evidence. This approach calls for the development of such evidence, and in particular for the expansion of controlled experimental studies of policing practices (Sherman, 1998).
 

Categorizing Recent Police Innovations
Moore, Sparrow, and Spelman (1997) suggest four distinct categories of police innovation: programmatic, administrative, technological, and strategic. These categories are not clearly separated from each other and, as Moore and his colleagues admit (1997), assigning any one innovation to one category over another is often a judgment call. Programmatic innovations establish new operational methods of using the resources of an organization to achieve particular results. These programs can include arresting fences as a way to discourage burglary, using police officers to provide drug education in the schools, and offering victim-resistance training to women. Administrative innovations are changes in how police organizations prepare themselves to conduct operations or account for their achievements. These include new ways of measuring the performance of an individual officer or the overall department as well as changes in personnel policies and practices such as new recruiting techniques, new training approaches, and new supervisory relations. Technological innovations depend on the acquisition or use of some new piece of capital equipment such as nonlethal weapons, DNA typing, or crime mapping software.
Strategic innovations represent a fundamental change in the overall philosophy and orientation of the organization (Moore, Sparrow, and Spelman, 1997). These changes involve important redefinitions of the primary objectives of policing, the range of services and activities supplied by police departments, the means through which police officers achieve their goals, and the key internal and external relationships that are developed and maintained by the police. Strategic innovations include shifting from ‘law enforcement’ to ‘problem solving’ as a means of resolving incidents, forming working relationships with community groups as a tactic in dealing with drug markets, and recognizing citizen satisfaction as an important performance measure. These innovations are strategic because they involve changing some of the basic understandings about the ends or means of policing or the key structures of accountability that shaped overall police efforts under the standard model of policing (Moore, Sparrow, and Spelman, 1997). We feel that the eight innovations described in this volume represent related attempts to change the ends and means of policing and, therefore, should be regarded as strategic innovations.
Weisburd and Eck (2004) suggest that recent strategic innovations expand policing beyond standard practices along two dimensions: diversity of approaches and level of focus (see Figure 1). The ‘diversity of approaches’ dimension represents the content of the practices employed or tools used by the police. As represented by the vertical axis, tools can range from mostly traditional law enforcement to a wide array of approaches. The horizontal axis represents the extent to which police practices are focused or targeted. Weisburd and Eck (2004) contrast standard police practices with hot spots policing, problem-oriented policing, and community policing. The standard model of policing, with its emphasis on enforcing the law and its generalized application of law enforcement powers, scores low on both dimensions. Hot spots policing scores high on focus, but low on the diversity of tools used to control hot spot locations. Problem-oriented policing rates high on diversity of tools and focus as the approach challenges police officers to implement strategies designed to deal with the underlying conditions that give rise to discrete crime problems. Community policing, where police draw on a wider array of resources to prevent crime and engage the community in defining and dealing with problems, scores high on diversity of approaches. However, when implemented without problem-oriented policing, the approach is not well focused on crime problems and provides a common set of services throughout a jurisdiction.
Another dimension that could be added to Weisburd and Eck’s (2004) classification of police practices is the degree to which the innovations change the goals of policing. Under the standard model, police departments were mostly focused on preventing serious crime by deterring and apprehending criminal offenders, serving justice by holding offenders accountable for their crimes, rendering immediate aid to people in crisis, and providing non-emergency services such as controlling traffic (Eck and Rosenbaum, 1994). While the eight innovations described identified above do not remove any of these goals from the tasks of policing, the new strategies rearrange the priorities among the goals and add new ones. Non-criminal and non-emergency quality of life problems receive much more attention from the new police strategies. Community and problem-oriented policing represent the most radical departures from standard police work. Community policing, in its various manifestations, challenges police officers to work with citizens to deal with a broader range of concerns, most notably fear of crime and social and physical disorder (Skogan, 2006). Problem-oriented policing similarly adds new goals to policing, but it also reorganizes police actions from focusing on incidents as units of work to focusing on classes of problems to be addressed by responses that can be quite different from routine police activities (Eck, 2006). Other innovations represent less dramatic changes to standard police goals. For example, disorder policing, if engaged without community and problem-oriented policing, expands the police mandate to include social and physical disorder but does not radically change the tactics engaged by the police to deal with these problems (Sousa and Kelling, 2006; Taylor, 2006).
 

Crime and Disorder Control Effectiveness
Research suggests that crime tends to cluster among a few places, offenders, and victims. For instance, Spelman and Eck (1989) examined several studies and estimated that 10 percent of the victims in the United States are involved in 40 percent of victimizations, 10 percent of offenders are involved in over 50 percent of crimes, and 10 percent of places are the sites of about 60 percent of crimes. In practice, the underlying conditions that give rise to crime problems and the resulting interventions to alleviate crime problems are likely to overlap these areas and, quite possibly, not fit nicely into distinct categories. For example, analysis of a gang violence problem may well reveal that much gang violence is retaliatory in nature (Decker, 1996) and that today’s offenders are tomorrow’s victims and vice versa. Analysis may also reveal that gang violence tends to cluster at particular places in the city (Block and Block, 1993; Kennedy, Piehl, and Braga, 1996). As such, police interventions to reduce gang violence may well address relevant features of places, offenders, and victims.
When police departments focus their efforts on identifiable risks, such as crime hot spots, repeat victims, and serious offenders, they are able to prevent crime and disorder (Braga, 2002; Eck, 2003). The strongest evidence comes from evaluations of hot spots policing initiatives (Weisburd and Braga, 2006; Skogan and Frydl, 2004). Braga (2001, 2005) presents evidence from five randomized controlled experiments and four quasi-experimental designs that hot spots policing programs generate crime control gains without significantly displacing crime to other locations. These crime prevention effects were reported at general crime hot spots (Sherman and Weisburd, 1995), high-activity violent crime places (Braga et al., 1999), gun violence hot spots (Sherman and Rogan, 1995), and street-level drug markets (Weisburd and Green, 1995). Moreover, in the five evaluations that examined immediate spatial displacement, hot spots policing initiatives were more likely to generate a ‘diffusion of crime control benefits’ to areas immediately surrounding the targeted hot spots (Clarke and Weisburd, 1994).
While the rigor of evaluation designs vary from simple before-after comparisons without control groups to randomized experiments, problem-oriented policing, when appropriately focused on specific crime problems, has been found to be effective in preventing crime (Sherman and Eck, 2002; Weisburd and Eck, 2004; Skogan and Frydl, 2004). Researchers have found problem-oriented policing to be effective in controlling a wide range of specific crime and disorder problems such as burglaries in apartment complexes (Eck and Spelman, 1987), prostitution (Matthews, 1990), convenience store robberies (Hunter and Jeffrey, 1992), and alcohol-related violence in pubs and clubs (Homel et al., 1997). The available scientific evidence on third party policing is derived from a similar mix of studies with varying degrees of rigor. Nonetheless, Mazerolle and Ransley (2006) report that third party policing is effective in dealing with drug problems, violent crime problems and problems involving young people.
Pulling levers strategies also seem to be promising in controlling the violent behavior of groups of chronic offenders (Braga et al., 2001; Wellford et al., 2005; Kennedy, 2006). While the pulling levers strategy has not yet been evaluated using the ‘gold standard’ randomized controlled experimental design, five quasi-experiments and four simple assessments have found violence prevention effects associated with the approach. Quasi-experimental evaluations in Boston (Braga et al., 2001), Chicago (Papachristos et al., 2006), Indianapolis (McGarrell et al., 2006), Lowell (Braga et al., 2006), Los Angeles (Tita et al., 2003) reported significant reductions in serious violence in the treatment cities and areas relative to comparison cities and areas. Simple pre-post assessments reported similar crime prevention gains in Baltimore (Braga, Kennedy, and Tita, 2002), Minneapolis (Kennedy and Braga, 1998), Stockton (CA) (Wakeling, 2003), and High Point (NC) (Coleman et al., 1999). This evidence provides an empirical basis for further program development, research, and evaluation (Welsh and Farrington, 2001).
As a general strategy, community policing has not been found to be effective in preventing crime (Mastrofski, 2006). The available research shows that unfocused community-oriented tactics such as foot patrol, storefront offices, newsletters, and community meetings do not reduce crime and disorder (Weisburd and Eck, 2004; Skogan and Frydl, 2004). However, as will be discussed below, there is strong evidence to suggest that community policing tactics reduce fear of crime (Weisburd and Eck, 2004; Skogan and Frydl, 2004). The available empirical evidence on the crime control effectiveness of broken windows policing is mixed (Sousa and Kelling, 2006; Taylor, 2006). It remains unclear whether police departments that engage a broad-based broken windows policing strategy actually reduce crime. Simple analyses of crime trend data suggest that cities experience decreases in crime after their police departments adopt Compstat (Silverman, 2006). However, since Compstat programs are often implemented in conjunction with other crime prevention initiatives such as broken windows and hot spots policing, it is very difficult to untangle the influence of Compstat on any observed crime control gains (Weisburd et al., 2006). Moreover, in New York City and three other cities, further analysis revealed the observed decreases in crime began before the implementation of Compstat (Eck and Maguire, 2000; Weisburd et al., 2003; Weisburd et al., 2006). Compstat has yet to be proven as an effective crime control strategy in cities that have adopted the approach.
Evidence-based policing has not been empirically tested as an overall model of policing (Welsh, 2006). However, evidence-based police departments would draw policies and practices from a solid research base of strategies that have proven to be effective in controlling crime (Sherman, 1998). While an evidence-based approach to policing may have the unintended effect of limiting the ability of police to innovate by privileging evidence over experience (Moore, 2006), we do not believe that engaging an evidence-based approach would undermine the crime and disorder control effectiveness of police departments.
 

Community Reaction to Innovative Police Strategies
In addition to concerns over the crime control effectiveness of the standard model of policing, police innovation in the 1980s and 1990s was also driven by high levels of community dissatisfaction with police services and a growing recognition that citizens had other concerns that required police action, such as fear of crime. Since citizen involvement in policing is a core element of community policing programs (Skogan, 2006), it is not surprising that we know most about citizen reaction to these types of programs. In general, broad-based community policing initiatives have been found to reduce fear of crime and improve the relationships between the police and the communities they serve (Weisburd and Eck, 2004; Skogan and Frydl, 2004).
Community policing strategies that entail direct involvement of citizens and police, such as police community stations, citizen contract patrol, and coordinated community policing, have been found to reduce fear of crime among individuals and decrease individual concern about crime in neighborhoods (Pate and Skogan, 1985; Wycoff and Skogan, 1986; Brown and Wycoff, 1987). Community policing also enhances police legitimacy. Citizen support and cooperation is closely linked to judgments about the legitimacy of the police (Tyler, 2004). When citizens view the police as legitimate legal authorities, they are more likely to cooperate and obey the law (Tyler, 1990). Public judgments about the legitimacy of the police are influenced by their assessments of the manner in which the police exercise their authority (Tyler, 1990, 2004). The available evidence suggests that the police generally obey the laws that limit their power (Skogan and Meares, 2004). However, minorities consistently express significantly lower confidence in the police when compared to whites (Tyler, 2004). Community policing improves citizens’ judgments of police actions (Skogan, 2006). For example, over an eight-year period of community policing, Chicago residents’ views of their police improved on measures of their effectiveness, responsiveness and demeanor (Skogan and Steiner, 2004). Importantly, these improvements were shared among Latinos, African-Americans and whites (Skogan and Steiner, 2004). Clearly, community policing has been a strategic innovation that has helped bridge the police confidence gap in minority communities.
While there is a growing body of systematic research on the effects of community policing on citizen satisfaction with the police, there is a noteworthy lack of research assessing the effects of other police innovations on police-community relations. This gap in knowledge is noteworthy as many of the contributions to this volume suggest a tension between the crime prevention effectiveness of focused police efforts and their potential harmful effects on police- community relations (Meares, 2006; Rosenbaum, 2006; Taylor, 2006; Braga and Winship, 2006; Weisburd and Braga, 2006; Mazerolle and Ransley, 2006). Certainly, legitimacy is linked to the ability of the police to prevent crime and keep neighbourhoods safe. However, the police also need public support and cooperation to be effective in preventing crime. While residents in neighbourhoods suffering from high levels of crime often demand higher levels of enforcement, they still want the police to be respectful and lawful in their crime control efforts (Tyler, 2004; Skogan and Meares, 2004). Residents don’t want family members, friends, and neighbours to be targeted unfairly by enforcement efforts or treated poorly by overaggressive police officers. If the public’s trust and confidence in the police is undermined, the ability of the police to prevent crime will be weakened by lawsuits, declining willingness to obey the law, and withdrawal from existing partnerships (Tyler, 1990; 2001). The political fallout from illegitimate police actions can seriously impede the ability of police departments to engage innovative crime control tactics.
This dilemma has been described elsewhere as ‘the trust dilemma’ (Altshuler and Behn, 1997). Innovation may be necessary for establishing public faith in the ability of government agencies to perform. But before the public grants government agencies a license to be truly innovative, it needs to be convinced that these same agencies have the ability to perform (Altshuler and Behn, 1997). Police departments should be encouraged to pursue effective strategies that aggressively focus on identifiable risks such as hot spots, repeat victims, and high- rate offenders. However, police departments must be careful in their application of these approaches to crime prevention. For example, anecdotal evidence suggests that ‘broken windows’ policing strategies enjoy broad community support as a legitimate way to reduce crime and disorder (Sousa and Kelling, 2006). However, when the ‘broken windows’ approach is distorted into so-called ‘zero-tolerance’ policing, indiscriminate and aggressive law enforcement can negatively affect police-community relations (Taylor, 2006). To avoid engaging tactics that will generate strong negative community reaction, police departments should encourage and embrace community involvement in their crime prevention efforts. In Boston, the involvement of black ministers in the police-led pulling levers violence prevention strategy allowed law enforcement agencies to pursue more intrusive and aggressive tactics that would not have been possible without community involvement (Braga and Winship, 2006).
 

Police Reaction to Innovative Strategies
The eight innovations differed in their degree of departure from the standard model of policing. The police most easily adopt innovations that require the least radical departures from their hierarchical paramilitary organizational structures, continue incident-driven and reactive strategies, and maintain police sovereignty over crime issues. In its most basic form, hot spots policing simply concentrates traditional enforcement activity at high crime places. The familiarity of the approach to police is straightforward as they have a long history of temporarily heightening enforcement levels in problem areas. While law enforcement tools are deployed in a new way, the pulling levers deterrence strategy focuses existing criminal justice activities on groups of chronic offenders. Broken windows policing involves making arrests of minor offenders to control disorder and, as an end product, reduce more serious crime. As Kennedy (2006) observes, ‘law enforcement likes enforcing the law.’ Strategies such as hot spots, broken windows, pulling levers policing appeal to law enforcement practitioners primarily because they allow mostly traditional tactics to be deployed in new ways with the promise of considerably greater results. Compstat, as implemented by most American police agencies, has been focused more on reinforcing and legitimating the traditional bureaucratic military model of police organization than on innovation in the practices of policing (Weisburd et al., 2003; Weisburd et al., 2006).
While all major American police agencies report some form of community policing as an important component of their operations (Bureau of Justice Statistics, 2003), the police have been generally resistant to its adoption. This is not surprising since community policing involves the most radical change to existing police organizations. Skogan (2006) and Mastrofski (2006) report many shortcomings in the practical application of its three core elements: citizen involvement, problem solving, and decentralization. Citizens are generally used as information sources rather than engaged as partners in producing public safety. Officers prefer law enforcement strategies to developing and implementing alternative problem-oriented responses. Most ‘community-oriented’ police agencies haven’t made the organizational changes necessary to decentralize decision-making authority to the neighbourhood level. Similarly, the available research on problem-oriented policing suggests that police officers experience difficulty during all stages of the problem-oriented process (Braga and Weisburd, 2006). Problem analysis is generally weak and implemented responses largely consist of traditional enforcement activities. Problem-oriented policing as practiced in the field is but a shallow version of the process recommended by Herman Goldstein (1990). Given its close relationship to community and problem-oriented policing, it seems likely that police departments engaging third-party policing will encounter similar practical problems.
It is not remarkable that the strategies that require the most radical changes to existing police practices and structures report the greatest difficulties in implementation. Nonetheless, the available evidence indicates a gradual transformation in police attitudes towards adopting these new strategies. In addition to the widespread reporting of innovative police practices across the United States, police officers views towards the community and problem-oriented policing philosophy are becoming more positive. As summarized by Skogan (2006), studies point to positive changes in officers’ views once they are involved in community policing, positive findings with respect to job satisfaction and views of the community, and growing support for community policing in districts that engage the strategy compared to districts that maintain traditional activities. Police history shows that it takes a long time for new models of policing to fully develop. The standard model of policing was itself a reform in reaction to corrupt and brutal police practices during the so-called ‘political era’ of policing (Walker, 1992). Initially, the reform movement progressed very slowly; in 1920, only a few departments could be labeled ‘professional’ or engaging the basic tenets of the standard model. It wasn’t until the 1950s that virtually all American police departments were organized around the principles set forth by O.W. Wilson, August Vollmer, and other reformers (Walker, 1992).
 

Conclusion: Police Innovation and the Future of Policing
What will the future bring? Will the police continue to innovate at a rapid pace? We don’t anticipate a new wave of strategic police innovation in the near future. The current context of policing suggests that future innovation will be incremental in nature. The conditions in the 1980s and 1990s that created the pressure for innovation simply no longer exist. Indeed, the atmosphere is precisely the opposite of earlier decades. Overall crime is down and federal funds available for demonstration projects to spur innovation are very limited. While the available research evidence is not as strong as some police executives believe, there is a general sense that these police innovations work in preventing crime and satisfying community concerns. This perspective on the crime control effectiveness of new police practices is reinforced by the modest research evidence briefly described here and in other reviews (e.g. Skogan and Frydl, 2004) and by a cursory examination of crime trends over the 15 years. The Federal Bureau of Investigation’s Uniform Crime Reports reveals a 33% decrease in the Index crime rate from 5,820 per 100,000 residents in 1990 to 3,899 per 100,000 residents in 2005 (http://www.fbi.gov/ucr/05cius/data/table_01.html). While no single factor, including innovative policing, can be invoked as the cause of the crime decline of the 1990s (Blumstein and Wallman, 2000), the ‘nothing works’ view of policing in the 1970s and 1980s (see, e.g. Gottfredson and Hirschi, 1990; Bayley, 1994) is no longer a topic of discussion in most policing circles.
It is important to note here that a recent increase in violent crime has generated concern among many urban mayors and police executives. In 2005, for the first since 1991, the Federal Bureau of Investigation (FBI) reported an increase in the violent crime rate.  Nationwide, homicides increased by 5% between 2004 and 2005. In smaller cities, homicides had increased by as much as 12.5%.  In media accounts, academics and practitioners have suggested that these recent increases in homicide are linked to a resurgence of urban gang violence and the availability of firearms (Johnson, 2006; Mansnerus, 2006). While violent crime rates and homicide counts are nowhere near the levels seen in the early years of the 1990s,  urban mayors and police executives are placing pressure on the Federal Government to reinvest in innovative police programs that could help cities address these concerning increases in violent crime (Police Executive Research Forum, 2006).
Many innovative changes to policing appear positive and have shown crime control and community benefits. However, this brief review reveals the need for greater research and knowledge about the effects of these innovations on police departments and the communities they serve. Relative to other criminal justice institutions, the police are very open to research and evaluation activities with universities and other research institutions (Skogan and Frydl, 2004). We believe that the police will continue to work with researchers to better understand crime problems, community concerns, police behaviour, and structural issues in their organizations. These collaborations will support the police in further refining their practices to become even more effective, fair, efficient, and accountable agencies in the 21st century (Skogan and Frydl, 2004).
Over the next couple of decades, we anticipate that individual police departments will continue to institutionalize innovative practices by making administrative adjustments to their organizations and by developing a set of supporting strategies that fit the nature of their crime problems in the neighbourhoods they police. Administrative arrangements and portfolios of crime prevention interventions will necessarily vary across departments as the police become more specialized in dealing with local crime problems. In essence, we believe police departments will continue their evolution from ‘production lines’ that engage a static set of processes that are used over and over again to produce the same result to ‘job shops’ where each police assignment is treated as a new challenge that might require a new solution (Moore, Sparrow, and Spelman, 1997).
We also anticipate some modest innovation in the development of systems to measure the performance of police departments. As police departments engage a broader set of tactics to deal with a wide range of community problems and concerns, there will be a need for sensible performance measurement that capture the value created by police along a number of dimensions such as reducing criminal victimization; calling offenders to account; reducing fear and enhancing personal security; guaranteeing safety in public spaces; using financial resources fairly, efficiently, and effectively; using force and authority fairly, efficiently, and effectively; and satisfying customer demands/ achieving legitimacy with those policed (Moore, 2002). Appropriate measurement plays a vital role in transforming police departments into the learning laboratories they are now positioned to become in the future (Maguire, 2004).
American police departments, however, will be challenged to maintain their current trajectory by the new set of homeland security demands created in the wake of the 9/11 tragedy. In many ways, this is a new crisis for police departments, as their goals will be further expanded by a new focus on preventing future terrorist attacks and dealing with potentially catastrophic events. On the one hand, this new set of demands, with its emphasis on collecting intelligence on terrorist networks, apprehending terror operatives, and protecting likely targets, may push policing back to a more professional model that is distant from the community. Indeed, there is real potential for a backward shift as federal financial support and attention has been directed toward enhancing local law enforcement’s role in maintaining homeland security while, at the same time, funding for community crime prevention efforts has been drastically reduced. On the other hand, this crisis may create a new source for innovation as police departments will strive to continue their recent success in dealing with crime and community concerns. The U.S. Department of Justice Community Oriented Policing Services office has already sponsored working group sessions and conferences on using community policing strategies to respond to the challenge of homeland security (U.S. Department of Justice, 2004).
Over the last two decades, the police industry has undergone radical changes in the ends and means of policing. This period of innovation has yielded a set of very promising strategies that can improve the ability of the police to prevent crime and enhance their relationships with the communities they serve. Police departments will be challenged to continue developing these new approaches while meeting the homeland security demands of the post-9/11 world.
Nonetheless, we believe that the future is promising for police agencies, as they will continue to evolve into more effective and legitimate governmental institutions.