Lagniappe Thoughts On The FBI'S Annual Report
Quick glances at clearance rates, firearm murders, CDE changes and more.
There was a ton of stuff in last week’s Crime in the US report that I’m still sifting through, and rather than spam everybody with a half-dozen posts in a week covering small-ish topics I’m going to hope to bundle a few together.
Clearance Rates
First, clearance rates1 rebounded in 2023 matching what I thought might happen based on state data. Clearance rates remain well below where they were pre-2020 for the most part though, and even then they weren’t particularly high.
The national murder clearance rate rose from 52 percent in 2022 to 58 percent in 2023 per the FBI. Remember that 2021’s data was incomplete, so those clearance rates are a guess based on less than full data.
I wrote about the record low clearance rates we were seeing in the FBI data last year and had a post earlier this year on what a pain it is to collect and analyze this data historically. So check those out if you want more info on this issue!
The outlier is auto thefts which probably reflects the continued surge in those crimes continuing through 2023. More crimes typically leads to a smaller share cleared. Only 8.2 percent of auto thefts were cleared in 2023 nationally, the lowest clearance rate ever recorded for any crime type in a given year. The motor vehicle theft clearance rate was 6 percent in New Orleans, 4.3 percent in Chicago, and 3.5 percent in Dallas.
The national violent crime clearance rate rose from less than 37 percent in 2022 to 41 percent in 2023. That's improvement, but still well below where they 45 percent of violent crimes that were reported cleared in 2019.
Firearm Murders
The share of murders committed with a firearm was roughly even in 2023 relative to 2022. Not all murders that are reported to the FBI come with information on what firearm was used, but 76.4 percent of murders involved a firearm. This share has remained largely steady over the last four years though it’s up considerably relative to where it was 10 years ago.
The share of firearms committed with a firearm is something of a guess based on the roughly 80ish percent of agencies that report murder characteristics each year.
My Guesses
My estimates were nearly exactly spot on! Check out what I wrote a few weeks ago below:
For those that don’t want to click a link, here’s what I guessed and why:
Murder - My guess was -10 percent and it was -11.6 percent. That’s close though I guess I should’ve trusted the numbers that were coming in from cities rather than peripheral measures like the CDC and Gun Violence Archive that suggested murder might be down a bit less than the cities were showing.
Violent Crime - My guess was -3.5 percent and it was -3 percent. I thought that the RTCI — which was showing a slightly smaller percent decline — would be understating the decline due to huge increases in robbery in DC and Chicago being outliers. That turned out to be accurate so go me!
Property Crime - My guess was -2 percent and it came in at -2.4 percent. This one seemed obvious analytically — big city auto thefts were skewing the decline more than they should’ve. Auto theft rose 12 percent nationally last year, but the big/small city split was clear from the RTCI and FBI quarterly data, so a bigger decline in property crime than RTCI predicted made sense.
The Quarterly Estimates Were Pretty Good
The FBI’s Q4 report for 2023 was published in March 2024 and it pointed to sizable declines in violent and property crimes with a historic decline in murder. The FBI’s annual estimates published last week showed slightly smaller declines in violent and property crimes with a historic decline in murder.
The Q4 report was off on average by 2.4 percent per crime type with the biggest miss being robbery which makes sense given Chicago not submitting quarterly data and having a huge increase in robbery last year.
Yes, the quarterly data is incomplete. And it has errors. And the quarterly trends are overstating declines by a fair amount earlier in the year. That said, the Q4 report released by the FBI accurately identified the nation’s crime trends six months before the full national estimates were released.
Accurate — albeit imprecise — and fast has value for correctly identifying the nation’s crime trends without having to wait an inordinate amount of time. Just a few years ago we would have had to wait until September of the following year without even having a solid understanding of those trends. Now we can generally know them before the year is out. That’s huge progress!
CDE Changes
The FBI made some improvements to the Crime Data Explorer including adding monthly crime counts. That’s great to have more readily accessible crime data, but people need to be very cautious when using this information because not every agency reports every month. NYPD, for example, reported crime counts quarterly to the FBI for about a decade and then didn’t report in 2021 due to the NIBRS switch.
If something looks weird in the crime data then it’s usually a reporting issue more times than not. So while having the monthly data easily accessible is good, it also carries user risk.
Participation
It's not always easy to tell which agencies reported data to the FBI each year. To help with this problem I made the below table of cities of 50,000 or more. The table shows each agency, population, whether they participated in UCR 2023, and whether they participated via NIBRS. Note that just because an agency participates doesn't mean they sent 12 months of data.
What’s Next
The FBI should release quarterly crime data for the second quarter of 2024 very soon and there is a lot more to evaluate from the CIUS 2023 report. I’ll take a (very cautious!) look at the hate crime data that was released last week soon, and it has been requested that I write about how the FBI estimates missing data, so I’ll do that as well.
The FBI defines clearance rates as “In the FBI’s Uniform Crime Reporting (UCR) Program, law enforcement agencies can clear, or “close,” offenses in one of two ways: by arrest or by exceptional means. Although an agency may administratively close a case, that does not necessarily mean that the agency can clear the offense for UCR purposes.”
The FBI just “revised” their crime statistics, without any announcement or fanfare explaining the changes.
The “new” data shows Medicare’s a 4.5% INCREASE in crime for 2022.
The 2023 agency participation is incredibly helpful - thank you! Where did you get these data? Sifting through all of the files on the CDE is confusing work.