Comment: Crime is down, but what else do the FBI’s stats show?
Published 1:30 am Monday, August 18, 2025
By Justin Fox / Bloomberg Opinion
Crime fell in the U.S. in 2024. This was not exactly news when the Federal Bureau of Investigation disclosed it in its annual crime report last week; the Major Cities Chiefs Association, Real-Time Crime Index and Council on Criminal Justice released data months ago showing almost every category of crime down in 2024 in the cities they track, and they have put out subsequent reports showing continued declines in 2025.
Still, it was nice to see the FBI’s annual report (1) arriving earlier than in past years and (2) again covering more than 95 percent of the U.S. population as it had before a botched switchover to a new reporting system led to a collapse in 2021 in the share of law-enforcement agencies that sent in data. That new National Incident-Based Reporting System (which is up to 87.2 percent population coverage, with about 2,000 mostly smaller agencies still using the old summary reporting system) is also beginning to deliver on its promise of providing much more detailed information about crime in the U.S.
With President Donald Trump portraying crime in Washington, D.C., and other cities as out of control despite the clear downward trend, and Gallup polls through the decades showing most Americans convinced that crime is rising no matter what the statistics say, it’s anybody’s guess what effect these new, more reliable and more detailed numbers will have. Even I am a little dubious of FBI estimates showing violent crime (murder, rape, robbery and aggravated assault) at half-century lows relative to population and property crime possibly lower than it’s ever been. The Bureau of Justice Statistics’ annual Crime Victimization Surveys show marked declines over the past two decades in the share of crimes that respondents say they reported to police, and those are the only crimes the FBI knows about; it even changed the name of its annual release from “Crime in the Nation” to “Reported Crimes in the Nation” this year to reflect this. Also, property offenses have been migrating online, rendering the historical FBI property-crime trio of burglary, larceny-theft and motor vehicle theft less relevant in an age of phishing, catfishing and other digital crime.
The new NIBRS system does make room for digital crime and lots of other offenses. It’s not much good yet for comparisons over time, but the snapshot it offers of crime in 2024 is enlightening. The biggest takeaway — which I’ll get to in a moment — is that crime in the U.S., at least violent crime, is mostly not of the random, impersonal sort that dominates headlines and nightmares but instead usually involves familiar people in familiar places doing terrible things to one another. The most fascinating information I found among the 2024 NIBRS spreadsheets, though, had no obvious takeaway other than: Watch out between midnight and 1 a.m.
I wasn’t able to find anything in the existing crime literature that explains this explosion in crime just after midnight. There’s evidence that darkness encourages certain crimes, but except perhaps in Alaska in summertime its no darker at 12:30 a.m. than at 11:30 p.m. There’s also much literature on the impact of bar closing times on criminal activity, but there are few places in the U.S. where last call is mandated to be as early as midnight. It is possible that midnight is when a lot of people who have been out on the town decide to head home, but I guess it’s also possible that it’s an especially memorable time marker that victims and police use when they’re not entirely sure when a crime happened. I look forward to the research aimed at sussing this out.
As you may notice from the above chart, NIBRS sorts crimes into baskets different from the violent-property divide used historically by the FBI. The crimes against persons basket overlaps with violent crime but doesn’t include robbery (which falls under crimes of property), does include kidnapping and human trafficking, and defines assault and sexual assault far more expansively than the aggravated assault and rape categories found in violent-crime estimates. The NIBRS assault offenses even include intimidation, which is how 26,700 such offenses in 2024 could be reported as having occurred in cyberspace. By far the most common location, though, was people’s homes, which accounted for 57.1 percent of the 4.1 million crimes against persons reported in 2024.
The 8.2 million property crimes reported in 2024 were much more dispersed, although residence/home was still the No. 1 location, with 33.5 percent of offenses, and highway/road/alley/street/sidewalk No. 2, at 12.1 percent. The 2 million crimes against society — drug offenses, pornography, prostitution, etc. — were more of a streets-and-sidewalks phenomenon, with 50.9 percent of reported offenses occurring there.
The overwhelming majority of crimes against persons were also committed by someone known to the victim.
I’ve included robbery here —which in the FBI definition is taking or trying to take something of value “under confrontational circumstances from the control, custody, or care of another person by force or threat of force or violence” — because of its historical status as a violent crime. Robberies are more likely to be committed by strangers. But the U.S. reported robbery rate is down 60 percent over the past two decides, which the relatively modest decline in the share of crimes that are reported comes nowhere near explaining. It is not a growing problem.
More concerning is what’s been going on with murder and assault. The national murder rate jumped to a 25-year high in 2020, and while it has retraced all of that increase, it was only slightly lower in 2024 than two decades earlier. The latter goes for aggravated assaults as well. By far the most common circumstances reported for murders and aggravated assaults in 2024 were arguments and domestic violence.
The big distinguisher between assaults and murders is the weapon used. For all assault offenses (not just the aggravated assaults; sorry, this is just how the FBI provides the data), hands, feet, teeth and other “personal weapons” are most common. For homicide offenses (which include negligent manslaughter as well as murder and nonnegligent manslaughter) it’s guns by a mile.
The NIBRS statistics also delve into crime by race, a deeply fraught topic. They show that Black Americans are overrepresented among known offenders of all the crimes tallied except extortion/blackmail, where their 13 percent share of offenders is a bit lower than their 13.7 percent share of the U.S. population.
Some of this is the direct product of stereotyping and racism; victims often identify offenders as Black even when they’re unsure, and police are more likely to stop and arrest members of racial minorities than white people engaged in similar behavior. But that’s not enough to explain all the disparity. Black Americans really do make up a wildly disproportionate share of homicide offenders, for example. Why am I confident in saying that? Because Black Americans are even more wildly overrepresented among homicide victims (48.8 percent compared with 47.2 percent of offenders), and we’ve already seen that most homicides are committed by acquaintances and family members.
Basically, Black Americans are more likely to be stuck in high-crime places, and especially places with high gun violence. They’re also much poorer on average than white or Asian Americans, which explains some of the gap between their share of property-crime offenders and property-crime victims.
Men don’t have those excuses yet were greatly overrepresented among criminal offenders in 2024.
Men’s 49 percent share of crime victims was more or less in line with their share of the population. The crimes where the offender-victim gap was biggest were the ones you’d expect: sex offenses, kidnapping, human trafficking. It might be a surprise to some that men are almost as overrepresented among homicide victims as among offenders (76.6 percent compared with 86.4 percent), but if you’ve read this far it really shouldn’t be a surprise to you. For all the attention understandably paid to mass shootings, random killings, assassinations, murders for hire and other headline-grabbers, homicides in the U.S. are mostly prosaic tragedies involving men with guns, especially younger Black men, getting into fights with one another.
I can think of some policies that might reduce the odds of this happening. Sending the National Guard into the streets of Washington isn’t really one of them.
Justin Fox is a Bloomberg Opinion columnist covering business, economics and other topics involving charts. A former editorial director of the Harvard Business Review, he is author of “The Myth of the Rational Market.”
