Read an Excerpt
The PerformanceStat Potential
A Leadership Strategy for Producing Results
By Robert D. Behn Brookings Institution Press
Copyright © 2014 Ash Center for Democratic Governance and Innovation
All rights reserved.
ISBN: 978-0-8157-2528-2
CHAPTER 1
CompStat and Its PerformanceStat Progeny
How and why was the original PerformanceStat leadership strategy adopted and adapted by many different public agencies and government jurisdictions?
Jack [Maple] is the smartest man I've ever met on crime.
WILLIAM BRATTON, Police Commissioner of New York City
Whenever anything is being accomplished, it is being done, I have learned, by a monomaniac with a mission.
PETER DRUCKER, Claremont Graduate University
One night in the winter of 1994, Jack Maple was sitting in Elaine's—an expensive, four-star restaurant and celebrity hangout on the Upper East Side of Manhattan—drinking. As befits any urban legend, the reports on what he was drinking differ. As Maple remembered the night, he was on his third glass of champagne. Maple's boss, William Bratton, recalled that he too was in Elaine's that night and insisted that Maple, because he knew that he "might get called to a crime scene at any time," usually drank multiple cups of double espresso.
Regardless of whether Maple was inspired by the grape or the bean, the dozen words that he scrawled on a napkin that evening have been engraved on the minds of numerous public executives who seek to improve performance and produce better results:
1. Accurate and timely intelligence
2. Rapid deployment
3. Effective tactics
4. Relentless follow-up and assessment
These are "the four principles" of CompStat, the leadership strategy developed in the New York City Police Department by Commissioner Bratton, Deputy Commissioner Maple, and their NYPD colleagues. Their purpose? To improve the department's performance, and thus to produce better results. Specifically, when Bratton became NYPD's commissioner, he committed himself and his organization to reducing the city's crime by 10 percent in the first year, 25 percent over two years, and 40 percent over three years.
The "CompStat Craze"
Indeed, in New York City, crime did drop. It dropped significantly. George Mason University's David Weisburd and Stephen Mastrofski and their colleagues have called CompStat "a major innovation in American policing." To George Kelling of the Manhattan Institute and William Sousa of Rutgers University, "Compstat was perhaps the single most important organizational/administrative innovation in policing during the latter half of the 20th century." William Walsh of the University of Louisville labeled it "an emerging police managerial paradigm." Dall Forsythe, now at New York University, called it an "effective management innovation." Mark Moore of Harvard has described it as "an important administrative innovation in policing." Indeed, in 1996, Harvard University's Kennedy School of Government honored NYPD's CompStat with an Innovation in American Government Award.
As in any field of human endeavor, an innovation that proves successful (or merely appears to be successful) is disseminated through professional networks and informal channels to others who adopt and adapt it. By 1999, five years after Maple first scribbled down his four principles on Elaine's napkin, a third of the 445 police agencies in the United States with more than 100 sworn officers that responded to a survey reported that they had "implemented a CompStat-like program," and another quarter of these agencies said that they were planning to do so. This "Compstat Craze," as Christopher Swope called it in Governing magazine, was certainly a broad and rapid diffusion of this innovation throughout the policing profession.
Moreover, as police departments across the United States—and around the world—created their own versions of CompStat, they found Maple's four points worth saluting. In the Los Angeles Police Department, where Bratton served as police chief from 2002 to 2009, "the elements of CompStat consist of four distinct principles." To the Philadelphia Police Department, "the philosophy behind COMPSTAT is deceptively simple. It is based on four principles which have proven to be essential ingredients of an effective crime-fighting strategy." In Columbia, South Carolina, the police department introduced Maple's ideas about CompStat using almost exactly the same words: "four principles which have proven to be key ingredients of an effective crime-fighting strategy." In Escondido, California, the police department reported that "the CompStat process is based on four crime fighting strategies." The Minneapolis Police Department used a different name, CODEFOR (for Computer Optimized Deployment—Focus On Results), but still emphasized the same "four elements essential to crime control."
Big-city police departments are not, however, the only ones using the CompStat approach. Numerous departments in small cities and towns have also adopted this leadership strategy; they too have emphasized the importance of Maple's four principles. Sandy Springs, Georgia (population 100,000), simply called them "the elements." In West Vancouver, British Columbia (population 42,000), the police department listed them as "the four principles." To the police in Shawnee, Oklahoma (population 29,000), they are "four key principles." Burlington Township, New Jersey (population 20,000), called them "essential principles." Warwick Township, Pennsylvania (population 12,000), echoed Philadelphia, describing them as "four principles which have proved to be essential ingredients of an effective crime-fighting strategy."
Moreover, CompStat and Maple's four principles have been adopted not only to fight crime but also to reduce traffic accidents. And they have been employed not only by municipal police departments but also by police agencies at the state and provincial levels.
In the Canadian province of Manitoba, the Winnipeg Police Service created the term "CrimeStat"; the name may be different, but "the philosophy is built on [the same] four principles." In the State of Washington, the Highway Patrol named its approach "Accountability Driven Leadership," which "embraces many of the principles of COMPSTAT." In Australia, policing is the responsibility not of the municipalities but of the states, all of which created their own versions of CompStat (each with its own distinct name): Queensland named it Operational Performance Reviews; New South Wales, Operations Crime Reviews; Tasmania, Management Group Performance Reviews; Western Australia, Organisational Performance Reviews; and South Australia, Performance Outcomes Reviews. Only Victoria calls it "COMPSTAT." Not surprisingly, the Queensland Police Service reports that its Operational Performance Reviews are "based on the [same] four key elements."
In 1998 the NYPD created TrafficStat for the purpose of "reducing crashes, injuries, and fatalities, and increasing public safety." Others did the same. Troop C of the Louisiana State Police, based in Houma, created its own TrafficStat "composed of four components." In Canada, one of the components of the traffic safety program of the Ontario Provincial Police is "result driven policing," which is based on principles that include Maple's four.
For all of these police agencies, the purpose of creating their own version of CompStat is the same as the NYPD's: To improve performance and thus to produce better results, specifically, to reduce crime (and traffic accidents).
How, however, do Maple's four principles do that? How could they do that?
"Compstatmania": From CompStat to AgencyStat
Police departments were not the only public agencies to adopt CompStat. In what the Gotham Gazette called "Compstatmania," other public agencies in New York City—public agencies that had no responsibility for crime or traffic—also sought to adapt the NYPD's leadership strategy. Obviously, however, each agency had to adapt the concept to its own needs to improve its performance and thus to produce better results (whatever that agency's specific results might be). The New York City Department of Parks and Recreation launched ParkStat. The Human Resources Administration created JobStat, as well as VendorStat, Homecare VendorStat, HASAStat (for HIV/AIDS Services Administration), and MAPStat (for Medicaid). The city's Administration for Children's Services established ChildStat. Even the Off-Track Betting Corporation created BET-STAT. Honest!
Not all of these "children of CompStat," as the Gotham Gazette labeled them, had the suffix "Stat" in their name. For example, the city's Department of Correction created TEAMS, which stands for Total Efficiency Accountability Management System. Essentially, this is CorrectionStat. Similarly, the Department of Probation created STARS, for Statistical Tracking, Analysis and Reporting, or what might be called ProbationStat. The city's Housing Authority created APTS, for Authority Productivity Tracking System; think of this as HousingStat. The Department of Transportation called its version DOTMOVE.
Each of these New York City departments was seeking to improve performance—and thus achieve better results. Each of these public agencies sought to adapt the CompStat leadership strategy to its own purposes. Now the city didn't just have policing's CompStat. A variety of different city agencies had their own "AgencyStat."
During the years that Bratton and Maple developed CompStat, Rudolph Giuliani was New York City's mayor. Thus it is not surprising that when Giuliani ran for president in 2008, he promised to create CompStat-like programs in the federal government—FedStat, BorderStat, TerrorStat, and GAPStat. In the process, Giuliani invoked Maple's principles: CompStat, he told the press, "is all about: accurate and timely intelligence, effective tactics, rapid deployment of personnel and resources, and relentless follow-up and assessment."
Actually, several units of the U.S. government had already created their own AgencyStats. For the San Diego district of the U.S. Border Patrol, it was, indeed, BorderStat. And the Environmental Protection Agency called its version EPAStat.
Then, in December 2010, Congress passed and President Barack Obama signed the GPRA Modernization Act. The original GPRA, the 1993 Government Performance and Results Act, had been designed to "improve Federal program effectiveness" and to "help Federal managers to improve service delivery." Yet, it had not produced much improvement. Consequently, the GPRA Modernization Act, often referred to as the GPRA Mod Act, required federal agencies to conduct "quarterly priority performance reviews."
Even before the 2010 legislation the Food and Drug Administration had developed FDA-Track, the Department of Housing and Urban Development had established HUDStat, the Federal Emergency Management Agency was in the process of launching FEMAStat, and the Department of the Treasury was conducting quarterly performance reviews (though without a clever name). Now, Congress had apparently instructed all the agencies of the U.S. government to create something similar to NYPD's CompStat.
Yet could CompStat help the FDA or HUD or FEMA produce their quite different results?
CitiStat: From CompStat to JurisdictionStat
In the fall of 1999, Martin O'Malley, a city councilor in Baltimore, was campaigning for mayor in a city with a lot of problems. From 1990 to 2000, the city had lost almost 85,000 residents, nearly one-eighth of its 1990 population. More than one in five residents lived below the poverty level, and 14 percent of the housing units were vacant. Moreover, Baltimore had a notoriously high crime rate.
Thus, O'Malley asked Jack Maple to help him create CompStat in Baltimore. But as they drove around the city, Maple kept asking why they couldn't apply the CompStat leadership strategy to other aspects of city government. O'Malley demurred, but Maple kept insisting. The result was CitiStat, the first effort to adapt this leadership strategy to improve the performance of an entire governmental jurisdiction. And Maple's four principles were central to how O'Malley's leadership team thought about and created CitiStat.
Indeed, O'Malley remembers, "ComStat tenets immediately became our four basic tenets of CitiStat." In Baltimore, these official "tenets" of CitiStat—which has been continued and expanded by O'Malley's two successors as mayor, Sheila Dixon and Stephanie Rawlings-Blake—are quite similar to Maple's original four points. A decade after its creation, the "CitiStat Tenets" were on the wall of the CitiStat room and the CitiStat website:
—Accurate and Timely Intelligence Shared by All
—Rapid Deployment of Resources
—Effective Tactics and Strategies
—Relentless Follow-up and Assessment
Under Dixon, Baltimore explicitly recognized its debt to both Maple and CompStat:
CitiStat's Tenets were developed from the tenets created by Jack Maple for New York City's ComStat—a strategy that uses timely and accurate crime data to inform policing efforts. CitiStat uses the same tenets to provide timely, reliable services to Baltimore's residents.
And Rawlings-Blake continued the acknowledgment: "The CitiStat Office follows four main tenets. Those tenets were originally developed by Jack Maple."
Indeed, in the early years of the twenty-first century, CitiStat became the hot new thing. Journalists described it as a "Baltimore success story," a "pioneering innovation in across-the-board, eye-on-the-sparrow management," a "great success," "acclaimed," "much-emulated," a program that "may represent the most significant local government management innovation of this decade." In 2004, CitiStat won, again from Harvard's Kennedy School, one of the awards for Innovation in American Government.
In Baltimore, CitiStat has become an institution. It is how Mayors O'Malley, Dixon, and Rawlings-Blake ran the city. CitiStat was not a nice little add-on. It became central to their management of city government. CitiStat was the leadership strategy that these three mayors employed to ensure that city agencies were focused on their priorities—on improving specific aspects of their performance.
CompStat had not only evolved into AgencyStat. It had also evolved into JurisdictionStat. Yet exactly how could Maple's four principles improve the performance of an entire government?
The "Pilgrimages" of "Urbanistas"
Then, just as police chiefs had flocked to watch NYPD's CompStat, reported Charles Bens of Best Practices Consulting, "mayors began to show up in Baltimore to learn more about the CitiStat miracle." Neal Peirce, a columnist with the Washington Post Writers Group, noted that "the CitiStat model is spreading." Observed Sam Allis, a columnist for the Boston Globe, "Mayors and urbanistas from across the country visit Baltimore, and most return home believers." O'Malley's "urban innovations—primarily CitiStat," reported Time magazine, "have brought other curious mayors on pilgrimages to Baltimore."
Many of these believers created their own version of CitiStat. Primarily they did so in large cities with the capacity to develop an analytical staff to organize and examine "accurate and timely intelligence" (table 1-1). Nevertheless, several small municipalities also decided they could make effective use of this leadership approach (table 1-2).
Moreover, at least three counties—Montgomery and Prince George's Counties in Maryland and Cuyahoga County in Ohio—have created their CountyStat. In addition, King County, Washington, established its own KingStat, while neighboring Snohomish County created SnoStat. The CitiStat leadership strategy has even spread beyond the borders of the United States: The London Borough of Barnet created FirstStat, and in the Netherlands, Rotterdam created MaasStat (named for the Maas River, which flows through the city).
(Continues...)
Excerpted from The PerformanceStat Potential by Robert D. Behn. Copyright © 2014 Ash Center for Democratic Governance and Innovation. Excerpted by permission of Brookings Institution Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
