The crime analysis and management approach known as COMPSTAT is widely heralded and frequently imitated. The tributes describe COMPSTAT as ”perhaps the single most important organizational/ administrative innovation in policing during the latter half of the 20th century” (Kelling and Sousa 2001,6). A Criminology and Public Policy Journal editor termed COMPSTAT ”arguably one of the most significant strategic innovations in policing in the last couple of decades” (Criminology and Public Policy 2003, 419). The authors of a major study note that COMPSTAT ”has already been recognized as a major innovation in American policing” (Weisburd et al. 2003, 422). In 1996, COMPSTAT was awarded the prestigious Innovations in American Government Award from the Ford Foundation and the John F. Kennedy School of Government at Harvard University. Former New York City Mayor Rudy Giuliani proclaims COMPSTAT as his administration’s ”crown jewel” (Giuliani 2002, 7).
Because most people only observe its most visible elements, COMPSTAT is frequently misunderstood. These visible elements include up-to-date computerized crime data, crime analysis, and advanced crime mapping, which serve as the bases for regularized, interactive crime strategy meetings in which managers are held accountable for specific crime strategies and solutions in their areas. It is fair to say that the widespread diffusion of COMPSTAT refers to these most noticeable elements. Since COMPSTAT was first unveiled by the New York City Police Department (NYPD) in 1994, a Police Foundation’s 1999 survey for the National Institute of Justice (NIJ) revealed that a third of the nation’s 515 largest police departments had implemented a COMPSTAT-like program by 2001 and 20% were planning to do so. The same survey found that about 70% of police departments with COMP-STAT programs reported attending a NYPD COMPSTAT meeting (Weisburd et al. 2001).
This process has continued in subsequent years. Gootman (2000, B1) reported that 219 police agency representatives visited NYPD COMPSTAT meetings in 1998, 221 in 1999, and 235 in the first ten months of 2000. Attendance at a COMPSTAT meeting, while a useful introduction, does not provide adequate preparation for introducing and establishing COMPSTAT into one’s own department. In fact, the presentations at such meetings may be misleading because attendees often become mesmerized by the flashy overhead display of multiple crime maps synchronized with technologically advanced portrayals of computerized crime statistics. In the vernacular, it is only necessary to display computer-generated crime maps and pressure commanders ”to make the dots go away” (Maple 2000, 38). This more elaborate but superficial approach is emblematic of the quick managerial fix approach, thus contributing to the misunderstanding and misapplication of COMPSTAT.
Performance Management Comes to Policing
More importantly, COMPSTAT is, in many ways, emblematic of the application of business-oriented managerial reforms to modern policing. Enhancing accountability for performance by adopting business and professional practices has repeatedly emerged as the holy grail of police managerial reform. Early twentieth-century police reformers such as O. W. Wilson sought to upgrade the quality of police performance through the introduction of sound business practices.
In recent years, the goal of refashioning police agencies so that they more closely mirror private organizations has been extended to entire police departments.
For example, Herman Goldstein’s path-breaking work advocated new managerial structures capable of reading and responding to internal and external work environments (Goldstein 1990, p.162).
By 1994, there was a clear trend within government to emulate business practices and ”reinvent” themselves by dramatically altering administrative structures and operational processes in order to enhance efficiency and the overall quality of performance. This all-embracing view is now shared by many police scholars urging police adoption of ”corporate strategies” and ”entrepreneurial” approaches (Burns and Stalker 1961; Moore and Trojanowicz 1988).
For police, as with other public organizations, performance measures were viewed as integral to the adoption of and accountability for these business characteristics (Kravchuk and Schack 1996; Wholey and Hatry 1992). The Government Performance and Results Act of 1993 (GPRA) and Osborne and Gaebler’s Reinventing Government (1992) promoted the reinventing government movement as the ”public sector analogue to the corporate ‘business process engineering’ movement, which has been described as ‘one of the most influential management ideas of the nineties”’ (Case 1999, 419, as cited in O’Connell 2002).
Performance measures are supported by concepts linked to the reengineering process. ”Benchmarking” and ”best practices” cherish objectives and standards that are shared throughout the entire organization (Bowerman and Ball 2000; Coe 1999).
The New York City Police Department
The introduction and spread of these business concepts and practices in American police departments is revealed through an examination of post-1993 COMPSTAT and related changes in the New York City Police Department (NYPD). Modern business management provided the orchestral score for these changes; reengi-neering was its name. Contemporary management literature explains that reen-gineering requires ”radical change,” a ”starting over” throughout the entire organization, nothing less than a ”reinvention of how organizations work.”
New York’s police commissioner at this time, William Bratton, was portrayed by the Economist as ”a fan of the re-engineering rhetoric of Michael Hammer and James Champy” (Economist 1995, 50). Business Week lauded the NYPD’s ”innovative turnaround artists” who used ”private-sector” techniques (Business Week 1995, 83). Each of twelve reengi-neering teams was dedicated to a specific topic and was asked to determine what was broken and how to fix it—or, perhaps more accurately, what would be used to replace it. Each committee was supplied with a copy of Hammer and Champy’s Reengineering the Corporation, which advocates ”abandoning outdated rules and fundamental assumptions” (Hammer and Champy 1993, 31).
The NYPD’s reengineering reports questioned the department’s operating procedures. What current policy yields, they claimed, was inadequate. The precinct organization report, for example, noted:
… 2 or 3% reduction in crime is not good enough. We need to change the organization to do more. The need to reengineer precincts is not immediately apparent. City-wide crime continues to decline year after year. Every annual precinct state of command report, without exception, includes evidence of neighborhood improvements. Bureau and Special Unit Commanders to a man, or a woman, will vigorously defend the effectiveness of the present system. Why fix what’s not broken?
The answer is in the new mission of the department to dramatically reduce crime, fear, and disorder. Slow, continuous improvement doesn’t cut it, and that is all the present system can deliver. . . . [R]e-engineering in its simplest form means starting all over, starting from scratch (New York City Police Department 1994, v).
COMPSTAT acted like a booster cable to the NYPD’s battery, providing the cranking power needed to activate decentralization and command accountability. Relinquishing control of daily ground operations was the most fundamental yet difficult challenge facing the new administration. Traditionally, the person at the apex of the NYPD pyramid would retain control through standardized procedures and policies. But in order to hold precinct commanders accountable for crime prevention, the new leadership knew the organization must grant them more discretion. Rather than allow headquarters to determine staffing and deployment on a citywide basis, it was decided that reducing crime, fear of crime, and disorder would flow from patrol borough and precinct coordination of selected enforcement efforts.
COMPSTAT emerged as the central mechanism with which up-to-date crime performance measures were developed, and precinct commanders were held accountable for crime in their areas. In the parlance of business management, performance standards were set for the whole department and its subunits. Entrance to Bratton’s higher echelon was restricted to commanders committed to double-digit crime reduction. Establishing a specific objective—a 10% reduction in crime for 1994—was the initial propellant for change. While target setting is the norm for the private sector, it usually is anathema for public organizations because it offers a yardstick against which performance can be more accurately measured and, if deficient, condemned.
The commissioner and his top aides recognized that data needed to be gathered and analyzed in a timely manner if effective crime reduction strategies were to be implemented. These statistics constituted the first COMPSTAT topic in February 1994. Subsequently, periodic meetings were scheduled at headquarters whereby precinct commanders were required to report and react to crime data generated from their areas of responsibility (that is, their commands). Over time, these data-based informal discussions between department executives and field commanders developed into formal biweekly strategy meetings (known as COMPSTAT meetings) at which all levels of the department participate to identify precinct and citywide crime trends, deploy resources, make assessments, and are held accountable for crime control strategies and results. (For details, see Silverman 1999.)
The first three years of the NYPD’s COMPSTAT program corresponded with dramatic declines in the city’s crime rate. According to the FBI’s Unified Crime Reports, the city’s 12% decline in index crime in 1994 (compared to a national drop of less than 2%) grew to 16% in 1995 and yielded another 16% in 1996. These decreases accounted for more than 60% of the national decline during this period.
Although these figures, of course, do not prove a causal relationship between COMPSTAT and a decrease in crime, they received extraordinary law enforcement and national attention. The New York model (COMPSTAT) was offered as the road to rapid crime reduction (Gootman 2000, B1). A Time magazine 1996 observation is still applicable ten years later: ”COMPSTAT has become the Lourdes of policing, drawing pilgrim cops from around the world . . . for a taste of New York’s magic” (Pooley 1996, 55-56).
Response to COMPSTAT Critiques
In a valuable national study, the Police Foundation identified COMPSTAT’s ”six key elements.” They are ”mission clarification, internal accountability, geographic organization of operational command, organizational flexibility, data driven problem identification and assessment, innovative problem solving tactics, and external information exchange” (Weisburd et al. 2003, 427).
The study found many of these key elements lacking in many police COMP-STAT programs. In their comparison of COMPSTAT and non-COMPSTAT agencies, the study concludes that the COMPSTAT agencies ”have opted for a model much heavier on control than on empowerment” (Weisburd et al. 2003, 448). Moreover, despite its virtues, the authors found that:
COMPSTAT agencies were largely indistinguishable from non-COMPSTAT agencies on measures that gauged geographic organization of command, organizationally flexibility, the time availability of data, and the selection and implementation of innovative strategies and tactics. . . . COMPSTAT departments are more reluctant to relinquish power that would decentralize some key elements of decision making geographically . . . enhance flexibility, and risk going outside of the standard tool kit of police tactics and strategies. The combined effect overall, whether or not intended, is to reinforce a traditional bureaucratic model of command and control. (Weisburd et al. 2003, 448)
This analysis fails to address the fact that the study’s COMPSTAT programs are self-designated. There is no evidence that these police agencies underwent the self-diagnosis, reengineering, and organizational and managerial overhaul processes that preceded the New York COMPSTAT experience. In reality, COMPSTAT is a revolution in thinking about the role and ability of the police to address crime as opposed to reacting to social and economic conditions that breed crime. COMPSTAT, then, is a performance management mind-set that some have likened to a paradigm change in policing. In accordance with classic bureaucratic structure, the overall orientation of managers within the department was ”downward,” rather than outward (toward the external environment) or upward. Precinct commanders ”did not see crime reduction as their foremost responsibility” and were ”essentially on their own in combating crime” (Silverman 1999, 98). Commissioner Bratton quickly altered this mindset by making a variety of high-level personnel changes and by redefining the department’s overall purpose and mission (Bratton 1998).
Both Henry and Wash follow Kuhn (1996) in their adoption of the ”paradigm” concept. Henry and Bratton (2002, 15) posit that paradigms are a sort of mindset or a collection of organizing principles and fundamental viewpoints around which we organize our basic understanding of the world. Paradigms can be compared to ideologies, belief systems, philosophical principles, or cognitive models that shape our understanding of something.
Walsh, in 2001, adopted this approach when he depicted COMPSTAT as representing an ”emerging police organizational management paradigm” (p. 1). Three years later, Walsh and Vito (2004) characterized COMPSTAT’s ”paradigm shift” as a ”goal-oriented, strategic-management process that uses information technology, operational strategy, and managerial accountability to guide police operations” (p. 57).
What started out as a computer file and a topic to satisfy crime informational needs has evolved and been reconstructed into a multifaceted forum for coordinated, reenergized, and accountable organizational crime fighting strategies. Its strength lies in its adaptability and compliance mechanisms. It is vitally important to recognize that COMPSTAT’s initial and prime raison d’etre was and is to measure and hold managers accountable for performance. In Moore’s words:
It becomes a powerful managerial system in part because the technical capacity of the system allows it to produce accurate information on important dimensions of performance at a level that coincides with a particular manager’s domain of responsibility. . . . [COMPSTAT] is, in the end, primarily a performance measurement system. (Moore 2003, 470, 472)
The Future of COMPSTAT
Today, COMPSTAT crime reduction efficacy is frequently advocated by police administrators, several of whom moved from the NYPD to head other city police departments. COMPSTAT’s introduction in New Orleans, for example, corresponded with a decline in murders from 421 in 1994, diving 55% in 1999 to 162. Minneapolis’s version of COMPSTAT, CODEFOR (Computer Optimized Deployment-Focus on Results), has been credited with a double-digit decrease in homicides, aggravated assaults, robberies, burglaries, and auto thefts between 1998 and 1999 (Anderson 2001, 4). In 2000 COMPSTAT was introduced in Baltimore by its new chief, a former NYPD deputy police commissioner. By the end of the year, there had been fewer than 300 homicides in Baltimore for the first time in twenty years as well as an overall crime drop of 25% (Anderson 2001, 4; Clines 2001, 15; Weissenstein 2003, 27). Between 1999 and 2001, Baltimore’s overall violent crime declined 24%, homicides dropped 15%, shootings fell 34%, robberies dropped 28%, rapes 20%, and assaults 21% (Henry and Bratton 2002, 307). Philadelphia’s former police commissioner, another former NYPD deputy police commissioner, attributed a decline in the city’s crime to COMPSTAT-driven policing. ”Social conditions in the city have not changed radically in the two years and we have the same police department, the same number of officers.
Nothing has changed but how we deploy them and utilize them” (Anderson 2001,3).
Similar crime reduction assertions have been made for police agencies around the world. Two Australian scholars, for example, recently published an evaluation of the New South Wales COMPSTAT-mod-eled Operation and Crime Review (OCM). The authors found that this process was effective in reducing three of the four offense categories studied (Chilvers and Weatherburn, 2004). Omaha’s year-old COMPSTAT is credited with improving cooperation among all units. ”Instead of one unit tackling a problem, everyone gets involved. . . . The entire culture has changed now” (Law Enforcement News 2004, 11).
In 1996, New York City’s Corrections Department modeled its TEAMS (Total Efficiency Accountability Management System) program on COMPSTAT with an examination of the department’s ”most fundamental practices and procedures” (O’Connell 2001, 17). Again, accountability is a major theme, which, when fused with more accurate and timely statistical reporting and analysis and inter-unit cooperation, has been credited with a dramatic reduction in inmate violence. Between 1995 and 1999, stabbings and slashing declined from 1,093 to 70 (Anderson 2001, 3).
TEAMS has evolved and now addresses more than just jail violence. Its accountability system has been continually expanded to retrieve and assess almost six hundred performance indicators addressing such issues as religious service attendance, maintenance work orders, health care, overtime, compliance with food service regulations, completed searches conducted, and the performance of personnel who have been the subject of the department’s civility tests (O’Connell and Straub 1999a, 1999b, 1999c).
COMPSTAT accountability mechanisms, long a staple of the private sector, have also become increasingly attractive to non-law enforcement public agencies.
There are numerous examples. The New York City Department of Parks and Recreation, for instance, developed its own version of COMPSTAT, calling it Park-stat. When parks officials visited NYPD COMPSTAT meetings in 1997, they realized that they could utilize this system to develop and refine their Parks Inspection Program (PIP), which oversees the maintenance and operation of more than twenty-eight thousand acres of property throughout New York City. Now Parks Department data analysis and managerial accountability are combined with monthly meetings to assess overall conditions; cleanliness of structural features such as benches, fences, sidewalks, and play equipment; and landscape features such as trees, athletic fields, and water bodies (O’Connell 2001, 20). The percentage of parks rated acceptably clean and safe increased from 47% in 1993 to 86% in 2001 (Webber and Robinson 2003, 3).
One observer’s assessment of Parkstat ranks it comparable to COMPSTAT’s high rating:
The Parkstat program is continually developing. Indeed, the department recently renamed it Parkstat Plus and has expanded it to include a broader range of performance measures . . . relating to personnel, vehicle maintenance, resource allocation and enforcement activity to ensure superior service delivery. Parkstat stands as an excellent example of how the COMPSTAT model can be adopted and successfully implemented outside the field of criminal justice. (O’ Connell 2001, 21, 22)
Perhaps the most ambitious extension of COMPSTAT’s managerial accountability and informational exchange processes began in the city of Baltimore in mid-2000 when its mayor was delighted with the results of the Baltimore Police Department’s first year with COMPSTAT. Baltimore’s program, called Citistat (first developed by the COMPSTAT architect, the late NYPD Deputy Commissioner Maple), is an attempt to evaluate and coordinate performance on a citywide basis whereby supervisors report every two weeks (as opposed to the previous quarterly basis) on their departments’ performances. Citistat’s timely data permit the assessment and coordination of diverse social services dealing with graffiti, abandoned vehicles, vacant housing, lead paint abatement, urban blight, and drug use and drug treatment. Discussions are based on up-to-date information. Citistat meetings are similar to those of COMP-STAT whereby data, graphs, and maps are projected to track and display department performance.
So far, the city is pleased with Citistat’s development. There has been a 40% reduction in payroll overtime with a savings of more than $15 million over two years. Its director of operations maintains that ”The charts, maps and pictures tell a story of performance, and those managers are held accountable” (Webber and Robinson 2003, 4). The fact that the prestigious Innovations in American Government Award was awarded to Baltimore’s Citi-stat ten years after NYPD’s COMPSTAT received the same award speaks to the enduring concepts embedded in this managerial and organizational approach. Speaking for an audience at Harvard University’s Kennedy School, Stephen Goldsmith (2004) stated, ”Citistat is a management tool for public officials that translates into real, tangible results for citizens. Government leaders across the country and around the world are taking notice of its success—and for good reason.”
Citistat, like most COMPSTAT-type programs, seeks to lower the informational barriers that generally hinder intra- and interagency collaboration. Baltimore is constantly expanding the number of agencies included in the data analyses and its Citistat meetings. It appears that Citistat is the ultimate test of COMPSTAT’s ability to serve as the informational cement of reform, the central mechanism that provides communication links to traditionally isolated specialized units. Fragmentation plagues many organizations. Harvard management expert Rosabeth Kanter (1983, 301) calls it ”segmentalism” and notes, ”The failure of many organization-change efforts has more to do with the lack . . . of an integrating, institutionalizing mechanism than with inherent problems in an innovation itself.” Without COMP-STAT, fragmentation would continue to rule supreme. COMPSTAT’s confrontation of informational splintering can be indispensable to organizational well-being. COMPSTAT can serve as the organizational glue that bonds many changes together.
COMPSTAT lends itself to a variety of law enforcement and non-law enforcement contexts. Numerous additional agencies are currently adopting their own versions of COMPSTAT. These include New York’s Office of Health Insurance with its Healthstat, which is designed to assist uninsured New Yorkers enroll in publicly funded health insurance programs. ”In the first 18 months, participating agencies enrolled about 340,000 eligible New Yorkers” (Webber and Robinson 2003, 4). The Department of Transportation instituted MOVE, an accountability and performance management system that meets twice a month to assess operational performance. The cities of Miami, Pittsburgh, and New York are pursuing comprehensive COMPSTAT-like systems similar to Baltimore’s Citistat.
Societal needs for information sharing, data analysis, and effective organizational-managerial performance will only continue to proliferate and even expand COMPSTAT’s rapid diffusion. Two and a half years before the terrorist attacks of September 11, 2001, on the United States, COMPSTAT’s architect, the late Maple, was asked about the future of COMPSTAT. He replied:
This should not be limited to the police department. It should involve every city agency, the fire department, the building department, the transportation department; everybody should be contributing and coordinating. And other law enforcement agencies need to participate fully. The FBI, DEA and ATF offices in a city should be running their own number and then bring those to COMPSTAT meetings at the police department. (Dussault 1999, 2)
Now in this post-9/11 era, there are numerous calls to overcome institutional barriers by federalizing the COMPSTAT process in order to combat terrorism.
The intelligence and accountability mechanism known as COMPSTAT is tailor-made for combating terrorism. Applying this mechanism to America’s new war, however, requires solving one of the most enduring problems in policing: turf jealousy, especially between the FBI and local law enforcement agencies.
The FBI’s antiterrorism efforts should be COMPSTATED in every city where the bureau operates. Where a Joint Terrorism Task Force (JTTF) exists, the commanders of the agencies should meet on a biweekly basis to interrogate task force members about the progress of their investigations. Where JTTFs do not exist, the FBI should assemble comparable meetings with all relevant agency heads. The new Fedstat meetings would have two purposes: to ensure that each ongoing investigation is being relentlessly and competently pursued, and to share intelligence (MacDonald 2001, 27).
