Whenever the media report on crime they tend to rely on official statistics supplied by the police. In the 1990s, police agencies took their production of reported crime and arrest statistics to a new level in an attempt to demonstrate that policing was both effective, and new “scientific” analyses of crime reports would make them even more effective.
By 2000, about 33% of American police agencies were implementing these new forms of statistical alchemy under names like Comp Stat, Citi Stat, Fast Track, and Powertrac. The new forms of statistical reporting resulted in a flurry of “scholarly” publications using the new data and analyses to tell us what was working in policing.
The problem is that most thinking criminologists had always been suspicious of police-produced data. With the advent of Comp Stat, and the other modes of analysis, scholarly skepticism should have been even greater. Previously, police departments made the crime rate increase or decrease at will depending on how they recorded crime, or if they recorded crime at all.
Now the opportunity for obfuscation was even greater, and police departments around the country seized the opportunity with unbridled enthusiasm.
• In 1996, the Atlanta police forgot to include 22,000 reported crimes in their data to help make the city look safer for the Summer Olympics.
• In 1998, my old friends in the Philadelphia police department faced a Department of Justice investigation for under-reporting crime. Police commanders admitted that the practice made politicians and department superiors happy and advanced their career prospects.
• In 2000, the Sex Crimes Unit in Philadelphia simply took no action on thousands of reported rapes and sexual assaults, driving the crime rate down and solving crime by deleting it.
• In 2003, the New Orleans police department downgraded hundreds of major felonies to “miscellaneous incidents.”
• In 2005, the Broward County Florida sheriff’s office systematically downgraded crimes, and began a practice of only charging suspects with one offense even if they had been arrested for multiple offenses. This was part of a campaign to convince local, small municipalities to abolish their own police departments and enter into a contract relationship with the sheriff’s department.
• In 2010, as part of a survey conducted by John Jay over half of the 309 retired NYPD officers interviewed, most of who had been precinct commanders when Comp Stat was initiated, admitted they routinely manipulated their precincts’ crime statistics. They reduced felonies to misdemeanors or simply didn’t report crimes at all to make their “effectiveness” look better than it was.
• In 2011, it was revealed that in Milwaukee police records clerks had been instructed to change computer codes to reduce the reported violent crime rate.
Creating phony crime statistics is easy and it is done repeatedly. The methods used are obvious:
-Crime reports aren’t recorded or filed.
-Felonies are reclassified as misdemeanors.
-Property values are downgraded so the crime is not considered a felony.
-A series of criminal acts are recorded as an isolated single event.
-Rapes are recorded as “inconclusive incidents.”
-Unsuccessful assaults with guns where the shooter missed are classified as “criminal mischief.”
-Domestic violence incidents are downgraded to less serious charges.
There is really nothing new in all of this. In the 1970s, Richard Nixon convinced Congress to pass in Omnibus Crime Bill as a “trial run” covering only Washington, D.C. Criminologists for the most part scoffed at the legislation, which most felt would do nothing to reduce crime.
When D.C. crime rates plummeted, Nixon took credit and Congress took the bait and extended the legislation nationwide. Of course, what happened was that crime had not declined at all. Police simply reported the value of property lost in almost every case of theft and larceny as $49, $1 short of the value which would have made those crimes felonies.
It is also obvious that police can make crime rise at will. In 2008, the Phoenix police department played on unfounded public fears about immigration and crime when they reported 358 kidnappings in the city. They claimed there was a kidnapping for ransom every night, and that these crimes would spread to every border town.
Right-wing commentators like Bill O’Reilly and Lou Dobbs began calling Phoenix the “kidnapping and murder capital of the United States.” The federal government chipped in $2.4 million in grants to fight the kidnapping menace. But, alas, when the Office of the Inspector General looked at the kidnapping numbers they found the real number to be about 190, and most of them were not “kidnappings for ransom.”
In the end it turned out that the violent crime rate in Phoenix had been plummeting, and it was one of the safest cities in the country. At the most basic level, virtually all officially produced crime data contains a considerable amount of error and outright fraud. We’ve known that for years.
What is shocking is when the media and police “scholars” use data they know to be tainted as a measure of either crime or law enforcement effectiveness. If we were recording that behavior, we could code it as “serial fraud.”
Gary Potter, PhD
Professor, School of Justice Studies
Eastern Kentucky University