Bad data from the FBI mislead about crime
S E E D E D C O N T E N T
The fourth quarter 2023 crime report from the FBI, the federal government's keeper of crime data, is unreliable at best and deceptive at worst.
The FBI's preliminary 2023 data show murder declined by 13.2% across the country and violent crime dropped 5.7% compared to 2022 levels. Various news headlines have reported the FBI's numbers unquestioningly, claiming murder is "plummeting" and violent crime "declined significantly" to pre-pandemic levels.
But these latest figures warrant skepticism, as we outline in a new report. In fact, violent crime is up substantially from 2019 levels, and last year's apparent drop is less significant than it appears.
Part of the problem is how police departments report offenses to the FBI. The FBI asked, then demanded, that law enforcement agencies "transition" away from the system they used for decades to a new, more detailed but onerous one. The 2021 mandate to use NIBRS to submit crime data proved a disaster as overstretched departments, especially in large cities, failed to reach compliance and thus did not submit data.
In 2019, 89% of agencies covering 97% of the population submitted data, but by 2021, that coverage plummeted to less than 63% of departments overseeing just 65% of the population. Chicago, Los Angeles, and New York City all failed to submit crime data. To increase participation, the FBI relaxed the NIBRS requirement in 2022, allowing agencies to report via the legacy system.
But many other cities, such as St. Louis, which had transitioned to the new method, still struggle to comply and submit partial or faulty data. The FBI compensates by relying more heavily on "estimation," or informed guesswork, to fill in the gaps and produce aggregated data.
That method of inferring offense totals is based on similar jurisdictions and past trends but is prone to error since it cannot compensate for local factors or events. For example, comparing Baltimore's 2015 homicide total to similar cities' trends would produce a skewed result. Baltimore, beset by riots and a police stand-down, saw murder rise 62% that year. In peer cities, murders rose in Cleveland only 15% and fell in Detroit by 1% and Memphis by 4%.
And the figures the agencies do report to the FBI do not match the agencies' publicly reported figures. For Baltimore, the FBI reported 225 murders in 2023, but the city reported 262 — which means the FBI left out 37 murders. In Milwaukee, the police department reported a 7% increase in robberies, but the FBI showed a 13% drop. Nashville's own data tallied more than 6,900 aggravated assaults in 2023, but the FBI counted only 5,941, leaving almost 1,000 of those offenses "missing." This trend is consistent across the board: While 2022's FBI city-level figures track the police's own data, the 2023 numbers consistently undercount offense totals. Any year-to-year comparison overstates decline.
Other measures of crime levels undermine, or at least muddle, the veracity of the FBI's data, which rely on "reported" offenses by victims and law enforcement themselves. The federal government's own victims' survey, which attempts to capture the gap between the number of actual offenses and the number reported to police, shows much higher offense rates than the FBI does. Moreover, a rising share of victims are failing to report their victimizations at all. In 2022, only 42% of violent crime victims and 33% of property crime victims bothered to report the crime to police.
That underreporting reduces the reliability of FBI numbers in measuring actual offense levels. For example, robbery offenses, which constitute roughly 25% of all violent crime by volume compared to 5% for murder, declined 18% between 2019 and 2022, according to the FBI, while the victim's survey suggests a 30% rise.
Another complicating factor is underreporting by the police themselves, who might be under pressure to "downcharge" offenses or dissuade the victims from reporting the crime at all. While the prevalence of underreporting by the police is hard to quantify, an investigation found that between 2005 and 2012, the Los Angeles Police Department erased thousands of crimes, mostly violent assaults, by reclassifying them as lesser offenses or not capturing them at all. The fuzzy math artificially reduced the city's crime rate by 7%. Any such malfeasance, when officials are under immense pressure to show progress in fighting crime, would inject bad data into the FBI's estimation model, only compounding its errors.
Our analysis of 40 jurisdictions that both reported data to the FBI and the Major Cities Chiefs Association, which collects data from the largest police departments, shows that homicide declined 10.2% across 40 major cities in 2023 since 2022, but the FBI reported a 12.8% decline in those same jurisdictions. Similarly, the FBI reported a 6.6% decline in violent crime since 2022, but the same cities reported only a 4.5% drop, with the FBI counting 3,200 more violent crimes in 2022 than the MCCA and 2,600 fewer in 2023 — a net discrepancy of almost 5,900 offenses. That gap conveniently results in a more significant drop in crime levels year to year.
In reality, violent crime is up substantially from 2019 levels. In big cities, murder is still elevated — up 23% since 2019 across all 70 cities tracked by the MCCA and up 18% according to a 32-city analysis by the nonprofit organization Council on Criminal Justice. For aggravated assaults, CCJ's 25-city sample found those up 8%, while the MCCA larger sample of cities reported a 26% increase over the same period.
To say crime is down is like descending from a tall peak and standing on a high bluff and saying you are closer to the ground — a true but misleading statement. Worse, the FBI's crime data serve as a poor altimeter to judge how high (or low) crime actually is.
Tags
Who is online
547 visitors
An appropriate analogy for inflation as well....
In which country? You are aware that inflation is a global problem, right?
oops.
oops?
As I said, the analogy is appropriate.
The FBI Uniform Crime Report's (UCR's) 'murder and nonnegligent manslaughter' data are so seriously flawed that the only figure they can provide criminologists is the number of deaths investigated by police as potentially criminal homicides, and the weapons use data reported by the UCR and the National Crime Survey (victimization data) are unreliable as well.
Believe your eye test, not the three letter acronym test. It’s all you need.
The only big citified people I know who feel safer today are here on NTers.
The rest? Not so much.