Sunday, November 15, 2009
News, In this Issue..., Politics, Atlanta

After Pennington

Experts familiar with Atlanta’s departing chief of police say the city should audit its crime reports

By Stephanie Ramage

When Atlanta Police Chief Richard Pennington leaves office early next year, the first thing the city will need to do is an audit of the police department’s crime statistics.

That is the consensus of experts interviewed by The Sunday Paper in Atlanta and in New Orleans, where Pennington was police chief before being recruited by Mayor Shirley Franklin. It is also the opinion of Atlanta mayoral candidate Kasim Reed, who has been calling for an audit of police reports since July.

Pennington’s “legacy” in New Orleans was the unchecked use of COMSTAT, a crime data tracking and mapping system hailed by its developers in the mid-1990s as a technological dream tool for police departments. It’s the same system Pennington implemented in Atlanta after he was hired in July 2002. The problem, say experts in both cities, isn’t Pennington or COMSTAT, but the combination of Pennington’s numbers-oriented leadership style and COMSTAT’s vulnerability to manipulation.

In Atlanta, as was the case in New Orleans, Pennington holds weekly meetings with his police zone commanders to review crime trends. These are called COBRA meetings. COBRA stands for Command Operation Briefing to Revitalize Atlanta, but many police officers refer to the COBRA meetings as “ass-chewing sessions.” During the sessions, crime numbers in the city’s police zones are considered and compared. The zone commanders who show a decrease in crime—on paper, at least—are praised, while those who show an increase in crime are berated.

COMSTAT generates a crime map based on the reports that are fed into it, and therein lies its vulnerability to distortion—distortion that New Orleans saw up close at the end of the Pennington years and continues to cause problems there.

With pressure to bring down the numbers emanating from Pennington onto the zone commanders—or, as they are called in New Orleans, the district commanders—a cascade of pressure resulted, says Rafael Goyeneche, president of the New Orleans Metropolitan Crime Commission. Majors chewed lieutenants about the necessity to reduce crime, the lieutenants chewed sergeants, and the sergeants chewed regular patrol officers. Consequently, more serious crimes like burglary were downgraded to less serious offenses like “damage to property.” Orders may have never actually been given to downgrade crimes, says Goyeneche, but the relentless focus on bringing down the numbers encouraged downgrading.

New Orleans conducted several audits, including one in 1997 before Pennington left, and one in 2003 that took part of the last year of his administration into consideration. In every instance, irregularities in crime reporting were found. Goyeneche, who says his office is still looking into the department Pennington left behind, thinks Atlanta should do its own audit after Pennington’s departure.

“If I were going to be the police chief of Atlanta, I would want to know what crime problems I am inheriting from my predecessor,” says Goyeneche.

Lou Arcangeli, who was demoted from deputy chief of the Atlanta Police Department in 1997 to the rank of captain after blowing the whistle on the department’s failure to properly report 498 robberies and 56 rapes, concurs. He points out that he found numbers-juggling before Pennington and the installation of COMSTAT in Atlanta. Any crime reporting system, he says, is vulnerable to manipulation if there is no outside scrutiny, and if leadership pays more attention to statistics than to neighborhood policing.

“Based on what I have heard from neighbors, cops, and reporters, Atlanta has got to do something to restore the community’s confidence in the police numbers,” says Arcangeli, who left the APD in 2002 and now teaches criminal justice at Georgia State University. “Any mayor that doesn’t do an audit of the reports is scripting failure, because how can real crime numbers match up to numbers that may have been deflated?”

Pennington responds with a statement e-mailed to The Sunday Paper: “The Central Records Unit is responsible for classifying our crime statistics in accordance with the guidelines of the FBI’s Uniform Crime Reporting. The department’s Tactical Crime Analysis Unit (TCAU) compares the management numbers against the numbers from the Central Records Unit to ensure accuracy. Each zone is responsible for analyzing crime statistics for their area of responsibility.”

Pennington continues: “At our weekly Cobra meeting, we discuss crime with the commanders who also have personnel looking at their crime reports and we compare them with the TCAU numbers. The accuracy of our crime statistics have been audited by the Federal government and also the Georgia Bureau of Investigation. Each audit has shown the Atlanta Police Department to be compliant in our uniform crime reporting practices.”

Arcangeli, however, says that when the feds and the GBI do an audit for the sake of the FBI’s Uniform Crime Report, they are merely counting incidents reported to the FBI against the reports submitted to the APD’s central records unit. They never see the reports that aren’t submitted, nor does their count make them aware of “downgrading”—changing an offense to something less serious.

“Manipulation of crime statistics starts when the police officer doesn’t make a report, or he discourages a citizen from making a report, or his commander interferes and has him change a report,” says Arcangeli.

The way to check, he says, is to interview the resident who originally placed the call to police.

 WHY DID WE RECRUIT PENNINGTON?

Both of Atlanta’s mayoral candidates, Mary Norwood and Kasim Reed, have said they would prefer to hire the next chief from within the APD, in large part because citizens and police officers have complained about Pennington's performance.

Reed, as head of the transition team charged with finding a police chief for newly elected Mayor Franklin in 2002, coordinated the efforts of executive search firms that had been hired for the task. The prestigious firm Spencer Stuart assembled the pool of candidates.

At that time, Pennington enjoyed a public approval rating in New Orleans of 85 percent. His popularity with the national press was off the charts. So it came as no surprise to anyone that Spencer Stuart put him forward as a best bet for Atlanta’s police chief.

An audit of crime reporting conducted by the City of New Orleans’ Office of Municipal Investigation in 1997 had been effectively muzzled by the New Orleans mayor’s office. Almost no one outside the city had heard about it.

Instead, the media, particularly the national media, kept singing Pennington’s praises, drowning out voices like that of Allen Johnson Jr., a reporter for the New Orleans alternative newspaper, Gambit, who wrote about the COMSTAT/Pennington problem in January 1998 and restated his findings in August 2002: “The pressure to keep the crime down, reinforced by weekly COMSTAT meetings at police headquarters, also led to allegations of under-reported crimes in a few of the city's eight police districts. The alleged objective, our police sources said, was to keep the heat off the district commander.”

But it was too late. Pennington was hired by Atlanta in July 2002.

“He had been featured on ’60 Minutes,’ six months before,” recalls Reed. “People don’t remember that in January 2002, when Shirley Franklin took office, the city of Atlanta was facing an $85 million budget shortfall. Absolutely every bit of energy and focus the transition team had was spent in dealing with that.”

So finding a police chief was left up to the search firm, Spencer Stuart, which forwarded Pennington’s resume to Franklin. On it, Pennington lists nearly 50 awards and honors bestowed on him by New Orleans civic and neighborhood organizations.

A New Orleans official who spoke with The Sunday Paper on condition of anonymity remembers “Some guy came in here from Atlanta to ask about him [Pennington]”—I ask him if it was a black guy named Kasim Reed, by any chance—“No, no, it was a white guy. I don’t remember his name. But when we brought up the COMSTAT stuff, he said ‘It’s not going to matter, the mayor has already made up her mind it’s going to be him.’”

Much of Pennington’s glowing press was based on one promise that he kept.

 THE UNTOUCHABLE

In New Orleans, Pennington inherited the highest murder rate and the most corrupt police department in the nation. It would have been difficult for him to disappoint anyone.

At about the same time that Pennington was being sworn in as chief of the New Orleans Police Department on Oct. 13, 1994, NOPD Officer Len Davis was urging drug dealer Paul Hardy to kill Kim Marie Groves, a woman who had filed a brutality complaint against Davis.

Federal Bureau of Investigation wiretaps of Davis’ phone revealed the cop impatiently pushing Hardy to “Come do the bitch now!”

No sooner was Pennington sworn in than the FBI pulled him aside and told him they estimated that between 15 and 20 percent of the NOPD’s officers were involved in some form of corruption. The murder of Groves was only the latest in a string of horrors.

Owing to the FBI’s keen interest in cracking down on the NOPD, Pennington was soon the recipient of more federal support than most police chiefs could ever dream of.

“He had enormous support from the Justice Department,” says longtime New Orleans civil rights attorney Mary Howell, who represents Groves’ survivors in a case against the city. “The FBI was here for years, so he was in a very unique situation here.”

All he really had to do was stay out of the way. He hadn’t been allowed to bring anyone with him from his old police department in Washington, D.C., so he was without confidantes at the NOPD, where officers routinely talked about feeding him to the alligators.

Naturally, the feds became his new best friends, and he became an object of adoration for New Orleans residents as he reformed their police force.

Clancy DuBos, part-owner and political columnist for Gambit, echoes Howell: “Generally he did get high marks as a reformer. We gave him high marks. New Orleans had a lot of bad cops and he got rid of them. That is not subject to debate. He really was a reformer.”

The clincher, says DuBos, was that Pennington kept his promise to cut New Orleans’ soaring murder rate in half.

“Homicide is a statistic you can’t fudge,” says DuBos. “It’s identified as a homicide by the coroner, not the police. And he did it—he cut the homicide rate in half.”

The payoff was excellent press, and residents so devoted that many of those who spoke with The Sunday Paper would take him back to this day.

A search on Pennington and his press throughout his tenure in New Orleans from 1994 until he came to Atlanta in 2002 is overwhelmingly positive. From the New York Times to the network news, Pennington was celebrated as the city’s savior.

“You have to understand, when he would walk into a restaurant, people would stand up and applaud,” says Howell.

THE PROBLEM WITH COMSTAT

But a problem was on its way.

In the mid-’90s, COMSTAT was sold to numerous police departments around the world by two former members of New York Police Chief Bill Bratton’s administration, Jack Maple and John Linder. (The Maple Linder Group would become Linder and Associates after Maple’s death in 2001.)

Used the right way, COMSTAT can help police figure out where they need to spend more resources. Used unethically, it provides police commanders with a unique opportunity for changing the crime rate in their areas with a click on a computer.

“It can be abused,” says Arcangeli. “As in any industry, if management expects certain numbers produced but doesn’t provide the support needed to produce those numbers, some people are going to cheat.”

Mayoral candidate Reed says he wants to do an audit because he has been struck over the past year by the disparity between citizen complaints he’s heard and reports that crime is down.

“It’s vital that citizens have confidence in the police department and that the data we are using to determine staffing, and where we should make investments, is accurate,” he says.

(Norwood did not respond to inquiries for this article.)

New Orleans was the first place Maple and Linder sold COMSTAT, and Pennington became an enthusiast.

Within the city of New Orleans, district commanders began to compete with each other to bring down crime numbers. Their bonus pay was at stake. Goyeneche explains that, as district commanders, captains were eligible for an above-rank pay grade. Their continued positions as district commanders depended upon their ability to bring down the crime numbers. If they couldn’t, they were reminded each week, they could be replaced.

Consequently, some of New Orleans’ decrease in crime was accomplished through “downgrading.” For example, when a resident’s home was broken into, but the burglar was startled before he could take anything, even though the kicked-in door was evidence of an attempted burglary, some NOPD cops were labeling such incidents “damage to property.”

Meanwhile, the media was reporting that crime was down, even though the citizens were increasingly uncomfortable with that assessment.

“You would hear that armed robberies were down,” says Howell. “And you would have people coming out of church and getting robbed.”

A New Orleans Times-Picayune editorial published in November 2003 noted that in six of the city’s seven districts, investigators “found errors in 10 to 25 percent of cases in which officers changed the original classification of an incident. To its credit, the 3rd District had no erroneous downgrades. The error rates in the remaining districts are unacceptably high. … A 2001 review by Deputy Superintendent Duane Johnson found that downgrades far outnumbered upgrades, and that documentation was missing in some cases and in other cases did not support the changes.”

In late 2002, “an Inspections Division investigation found that 12 ‘negligent injury’ cases from the 6th District should have been classified as aggravated batteries. What’s most disturbing is that it was district policy, officers told Capt. Dominick Bondi, to classify shootings as negligent injuries if the perpetrator’s intentions were unclear or if victims were unwilling to pursue the matter.”

Howell recalls that crime reporting became the butt of jokes throughout the city.

“People would say ‘I seem to have lost my microwave,’” she says. “There were a lot of ‘lost’ microwaves or TVs at the time.”

The idea was to decrease crime on paper. Downgrading allowed certain NOPD districts—which are comparable to Atlanta’s police zones—to keep the numbers down in the Part I section of the FBI’s Uniform Crime Report (UCR). Part I deals with the kinds of crime that scare people—robberies, assaults, etc. The UCR is the annual report that every city in America looks at to determine its crime rate and that of other cities competing for business, tourism, or prestige.

Goyeneche calls COMSTAT the “legacy of the Pennington era,” and while he concedes that it can be useful, he points out that the nexus of COMSTAT and the FBI’s UCR has had some unpleasant side effects. The UCR relies on crime statistics compiled by local police departments, but if the cops are manipulating their crime statistics, the UCR isn’t going be an accurate reflection of crime. And when the UCR is being used politically, the worst can be expected.

“The UCR gets used for things for which it was never intended,” says Goyeneche. “It gets used for confirmation that the mayor has followed through with his or her promises to reduce crime, and since the police chief serves at the pleasure of the mayor, the system is vulnerable to manipulation.”

If COMSTAT is used alone without outside auditing, it may foster an atmosphere in which minor offenses are pursued, while time-consuming investigations into serious crimes are avoided.

“This is a continuing, deep-seated problem,” says Goyeneche. “This is not unique to New Orleans. There have been scandals involving downgrading of crime statistics in New York City, Miami and Atlanta.”

Downgrading victimizes the public a second time, says Goyeneche. 

Because police manpower is allocated according to the areas of need shown by COMSTAT, if the numbers aren’t accurate, some neighborhoods that really need more policing don’t get it. It also puts officers in danger. He uses the example of a police officer who looks at the stats and thinks that he’s probably being called to a “damage to property” incident reported by a neighbor, only to walk in on an armed burglar.

Peter Scharf, a criminal justice professor at Tulane University, says COMSTAT was seen by some officers as an excuse for not doing the kind of neighborhood policing that would have allowed them to spot crime trends by simply talking to residents.

“Cops call that GOYAKOD,” says Scharf. “Get off your ass and knock on doors.” SP

EDITOR'S NOTE:  To read an interview with the recruiter who actually vetted Pennington, visit http://www.sundaypaper.com/Blogs/TheRamageReport/tabid/235/articleType/ArticleView/articleId/4721/Default.aspx