Former President Donald Trump said FBI data that show homicides and other violent crimes trending down are “fake numbers.” They’re not.
The FBI data for 2023 are preliminary, but crime statistics experts say the reporting behind the overall downward trend is solid, and that trend is validated when compared to data samples from local and state law enforcement reports.
The FBI statistics contradict Trump’s campaign narrative, repeated at a May 1 rally in Wisconsin, about rampant and rising violent crime in the U.S. And polls that show most Americans believe crime is on the increase. But that doesn’t mean the data are wrong or “fudged,” as Trump put it.
The FBI statistics are, however, incomplete, given that they measure only crimes reported to law enforcement — some crimes, such as rape, are historically greatly underreported — and not every law enforcement agency reports its statistics. That has been the case for decades.
Trump’s dismissal of the validity of the FBI’s crime statistics reminds us of when Trump was running for president in 2016 and falsely labeled the unemployment rates published by the Bureau of Labor Statistics as “phony numbers.”
At that time, Trump claimed the unemployment rate was really 42%. (The official unemployment rate then was 4.9%.) Less than two months after Trump took office, however, he was happy to cite BLS’ official unemployment rate. Sean Spicer, Trump’s top spokesman at the time, joked in a press briefing about Trump’s new embrace of the jobs data, “Yeah, I talked to the president prior to this, and he said to quote him very clearly — ‘They may have been phony in the past, but it’s very real now.’”
According to the FBI’s preliminary 2023 crime report, violent crimes dropped 5.7% between 2022 and 2023, and the number of murders declined 13.2%. That’s based on data from 79% of law enforcement agencies in the U.S., representing higher participation than any year during Trump’s presidency. Murders and violent crime went up in 2020, Trump’s last year in office, and murders continued to rise in 2021, albeit to a lesser extent, as violent crime ticked down a bit. Both have been declining since, according to FBI and other crime data.
Asked about the downward trend in violent crime figures in a Time interview conducted on April 12, Trump said he didn’t believe it, and he claimed, “The FBI gave fake numbers.”
Time National Politics Reporter Eric Cortellessa, April 12: Violent crime is going down throughout the country. There was a 6% drop in—
Trump: I don’t believe it.
Cortellessa: You don’t believe that?
Trump: Yeah, they’re fake numbers.
Cortellessa: You think so?
Trump: Well it came out last night. The FBI gave fake numbers.
Cortellessa: I didn’t see that, but the FBI said that there was a 13% drop [in homicides] in 2023.
Trump: I don’t believe it. No, it’s a lie. It’s fake news.
Cortellessa: Sir, these numbers are collected by state and local police departments across the country. Most of them support you. Are they wrong?
Trump: Yeah. Last night. Well, maybe, maybe not. The FBI fudged the numbers and other people fudged numbers. There is no way that crime went down over the last year. There’s no way because you have migrant crime. Are they adding migrant crime? Or do they consider that a different form of crime?
Cortellessa: So these local police departments are wrong?
Trump: I don’t believe it’s from the local police. What I saw was the FBI was giving false numbers.
We reached out to Trump’s press office for clarification about what Trump was referring to when he said, “it came out last night” that “[t]he FBI gave fake numbers.” We also asked for any other evidence to support Trump’s claim the FBI “fudged the numbers.” We got no response.
The FBI figures are based on voluntary reports by agencies nationwide. The final numbers and information about nationwide crime rates, which are adjusted for population, won’t be available until the FBI’s annual crime report is released in October.
“This data is preliminary and unaudited so agencies have time to submit data for additional months or fix apparent data errors (of which there are a handful), but it’s data coming from agencies themselves,” crime analyst Jeff Asher, co-founder of the New Orleans firm AH Datalytics, told us via email. “I like to think of those figures as accurate but not precise.”
In other words, he said, the drop in murders may end up being 10% or 11% lower in 2023 instead of 13.2%. And the drop in violent crime may be smaller than the 5.7% in the preliminary report. But, he said, the preliminary data “highlights the trend of rapidly declining murder and less rapidly declining violent crime.”
The downward homicide trend is backed up by AH Datalytics’ analysis of data about homicides from more than 200 large U.S. cities, which showed homicides declined by about 12% in 2023, Asher said. The FBI data also track with a large decline in shooting victims in 2023 documented by the Gun Violence Archives.
“Murder almost certainly declined at one of the fastest rates ever recorded in 2023,” Asher wrote in his 2023 analysis.
As for the downward trend in violent crime, that “is seemingly backed up by publicly available data from 14 states that published their data already showing a decline in violent crime in most states,” Asher said.
The Council on Criminal Justice’s crime report for 2023, published in January, found that homicides in 32 cities that provided such data were 10% lower—representing 515 fewer homicides—in 2023 than in 2022. The CCJ analysis also found there were 3% fewer reported aggravated assaults and 7% fewer gun assaults in 11 reporting cities, and 5% fewer carjackings in 10 reporting cities. The report found robberies and domestic violence incidents each rose 2% in 2023.
“Overall, crime rates are largely returning to pre-COVID levels as the nation distances itself from the height of the pandemic, but there are notable exceptions.” the CCJ report states. “While decreases in homicide in the study cities (and many other cities) are promising, the progress is uneven and other sources of crime information, including household surveys of violent victimization, indicate higher rates and more pronounced shifts than reports to law enforcement agencies.”
The latest figures from the Major Cities Chiefs Association also show a decline in murders and violent crime. The number of murders went down by 10.4% from 2022 to 2023 in 69 large U.S. cities that provided data, according to its report. Since 2020, murders in those cities have dropped by 8.6%. The latest report from MCCA shows violent crime continued to trend down in the first quarter of 2024, though homicides and other violent crimes remain above their pre-pandemic 2019 levels.
“Given the multitude of data sources pointing to the same widespread decline I’d say the FBI quarterly data is trustworthy in terms of the overarching trend while there still being a fair amount of uncertainty as to how large the declines in murder and violent crime may have been,” Asher said.
Richard Berk, emeritus professor of criminology and statistics at the University of Pennsylvania, agrees. “Violent crime generally appears to have been declining post COVID,” Berk told us via email.
“It’s clearly going down in the way it’s described” by the FBI, Berk said. And if you doubt that, he said, “If you use local data (e.g., from the Philly PD) you can bypass claims of FBI malfeasance. And if you do that, you get pretty much the same story.”
Nonetheless, he said, the process of tracking crime in the U.S. is “a very imperfect system.” For one, some violent crimes — such as rape and domestic violence — are far more underreported to police than others.
Crime Victimization Survey
The FBI’s Uniform Crime Report documents crimes reported to law enforcement. The government’s other national crime measure is the National Crime Victimization Survey, which estimates levels of various crimes based on a survey of about 240,000 people each year, asking whether they have been victims of various crimes. The two measures can vary, and have in recent years.
While the FBI Uniform Crime Reports were showing a decrease in violent crime between 2021 and 2022, the NCVS for 2022 — the latest year available — showed the serious violent crime victimization rate — which includes rape and sexual assault, robbery, and aggravated assault — rose from 5.6 the year before to 9.8 violent crimes per 1,000 population age 12 and older.
In an October report, criminologists at the Council on Criminal Justice wrote that the divergence between the FBI’s Uniform Crime Report and the NCVS “makes it uncertain whether violent crime actually went up or down in 2022.”
Nonetheless, that still doesn’t support Trump’s claim.
“These findings from the National Crime Victimization Survey show that the 2022 rate of nonfatal violent victimization increased compared to 2021, but was similar to the rate in 2018 and remained much lower than the highs of the early 1990s,” Kevin M. Scott, principal deputy director of the Bureau of Justice Statistics, said in a press release about the report for 2022.
Asher said there are numerous reasons the NCVS does not nullify the trends reported in the FBI’s Uniform Crime Report.
The NCVS “is terrific for outlining the contours of what does and doesn’t get reported but isn’t particularly great at measuring year-to-year trends,” Asher told us. In a post on April 8, Asher outlined some of the reasons for the discrepancy between the reports in 2022.
“The most obvious reason to avoid year-to-year direct comparisons is that UCR counts murder while NCVS does not,” Asher wrote. “Murder victims cannot be surveyed, so the reason for the crime’s absence makes sense in NCVS, but it’s also the crime that comes with the highest societal cost and I’m guessing it’s usually the crime that people are thinking about when they think about the nation’s violent crime rate. Murder is also the one crime that probably has decently accurate — albeit imperfect — counts each year.”
Murders have indisputably gone down in each of the last two years, after a spike in 2020 and a smaller uptick in 2021, though they are still a bit higher than 2019.
Asher also cites a lag time built into surveys that ask about crimes over the last six months, the fact that the surveys only include people 12 and older, and that surveys — by definition — have margins of error.
Asher also notes that the NCVS’ most recent survey is for 2022, while the preliminary FBI data is for 2023.
“In many ways, 2022’s violent crime trend isn’t particularly important relative to the direction implied in 2023’s preliminary reported crime trend sitting here in the spring of 2024,” Asher wrote.
In an op-ed for the Wall Street Journal on April 24, John Lott, an economist and president of the Crime Prevention Research Center, argued that the NCVS has revealed that violent crime is not down, just reporting of violent crime to police departments. He attributes that to large cities arresting fewer people, and thereby giving victims less incentive to report a crime.
“Law enforcement has collapsed in the U.S., particularly in big cities,” Lott wrote, and “many Americans [are] no longer confident that the legal system will protect them.”
Indeed, Ernesto Lopez, a research specialist at the Council on Criminal Justice, said the NCVS indicated that “non-reporting of aggravated assaults increased by about 29% from 2021 to 2022,” which he said, “could create an undercount of aggravated assaults.” Nonetheless, he said, “I generally would not classify the FBI data as inaccurate.”
Lott, whose controversial research on crime and guns is often cited by conservatives, also attributes the discrepancy between the 2022 FBI and NCVS data to low participation among local police departments that feed data to inform the FBI report. But participation rates in 2023 grew substantially.
A Change in Reporting Data
As we have written, starting in 2021, the FBI transitioned to a new system for local law enforcement agencies to submit data, requiring agencies to use what’s called the National Incident-Based Reporting System. That first year, only 60% of agencies reported crime data — police departments in the two largest U.S. cities, New York City and Los Angeles, were among the notable non-reporters — and so the FBI and the Bureau of Justice Statistics provided national estimates to fill in the gaps. Due to the low reporting level, some crime data experts cautioned not to make sweeping conclusions about crime trends.
Anna Harvey, a politics, data science and law professor at New York University, was among those who warned at the time that politicians were “kind of throwing around allegations and claims about crime that may or may not be accurate.”
In an email interview, Harvey told us participation rates have since improved dramatically.
The year the FBI transitioned to the NIBRS system for collecting crime data — 2021 — she said, only 60% of agencies reported. That increased to 71% in 2022, “which was better but still worse than pre-NIBRS rates,” Harvey said.
But the preliminary 2023 FBI report — the one that found a 13.2% drop in murders and a 5.7% decline in violent crimes in 2023 compared with 2022 — is based on 79.4% of agencies reporting.
“That’s pretty good!” Harvey said. “It’s higher than any reporting rate during the Trump presidency, and close to the highest observed reporting rate between 2000 and 2022 (81%).”
‘Migrant Crime’
Trump argued that the FBI crime statistics must have been “fudged” because they did not account for a wave of “migrant crime.”
“There is no way that crime went down over the last year,” Trump said in the Time interview. “There’s no way because you have migrant crime. Are they adding migrant crime? Or do they consider that a different form of crime?”
Crime data experts say Trump confuses how FBI crime data are collected and reported.
“The FBI UCR statistics do not track incidents such as illegal entries, failure to appear in hearings, etc.,” Lopez, of the Council on Criminal Justice, told us. “However, if a migrant commits an offense, such as a robbery, and that robbery is reported to the police, and that police department reports their crime incidents to the FBI, that incident will be reflected in official statistics.”
To be sure, there have been a number of high-profile crimes committed by immigrants in the country illegally this year, including the murder of nursing student Laken Riley in February and an assault on New York City police officers in January.
But Asher, of AH Datalytics, says there is no evidence in the data to indicate a migrant crime wave. Asher said that assuming a wave of crimes being committed by immigrants was too small to register in overall national trends of reported violent crime, he decided to analyze crime data to see if it was at least showing up along the U.S. border with Mexico. And so he looked at Texas crime data.
“Comparing violent crime rates in Texas border counties over time to violent crime in the US and statewide in Texas shows no evidence of increasing violent crime along the US border with Mexico,” Asher wrote. “The 14 counties along the Texas-Mexico border have seen a relatively steady violent crime rate below that of the rest of their state and the nation as a whole.”
“There are — and likely always will be — extraordinarily tragic individual incidents of crime to point to as anecdotal evidence of whatever wider trend one wants to assert,” Asher wrote. “But individual tragedies do not inherently constitute a crime wave, and the lack of an overarching surge in incidents shouldn’t detract from the tragedy of individual examples.
“Ultimately, the US crime data system is poorly set up to definitively answer the question of whether there is an immigrant-driven crime wave,” he said. “That said, the overall trend of declining violent crime nationally, and seeing no localized crime surges in the places I’d expect to see one if there was such a ‘wave’ strongly suggests that no such thing exists.”
A February New York Times analysis found that while 170,000 migrants have arrived in New York City since April 2022 — when Texas Gov. Greg Abbott began bussing migrants there to call attention to rising illegal immigration into his state — “the overall crime rate has stayed flat. And, in fact, many major categories of crime — including rape, murder and shootings — have decreased.”
Jeffrey Butts, director of the Research and Evaluation Center at the John Jay College of Criminal Justice, told the New York Times there was no evidence of a migrant crime wave.
“I would interpret a ‘wave’ to mean something significant, meaningful and a departure from the norm,” Butts said. “So far, what we have are individual incidents of crime.”
An ‘Imperfect’ System
The FBI’s Uniform Crime Reports are an “imperfect” way to measure crime in the U.S., Berk said. There isn’t uniform compliance among reporting agencies, and the report only captures crimes reported to law enforcement.
“Overall crime is a composite of many different kinds of crimes, some are quite common and some are quite rare,” Berk said. “Overall measures can be dominated by the most common crimes. Crimes such as homicide, which is what has great political clout, are relatively rare. The concept of overall crime is basically nonsense.”
“The best back of envelope way to proceed is to focus on particular crimes one by one and one jurisdiction at a time,” Berk said. “But there are subtitles here too. For example, homicides can fall even if the number of shootings increases insofar as medical care substantially improves, such as with the ‘scoop and run’ policy of the Philly PD. Trauma centers really help as well.
“Nevertheless, violent crime generally appears to have been declining post COVID,” Berk said.
In other words, there are numerous caveats that go along with crime statistics like the FBI’s Uniform Crime Reports. The data have certain limitations. But crime data experts say they are useful and informative, and there is no reason to believe they are “fake” or “fudged,” as Trump claimed.
In order to check the FBI data, Berk said, simply look at the crime reports from various cities and you will see they generally match up with the data reported by the FBI for that city, he said.
“It’s hard to argue there’s a conspiracy [by the FBI to fudge the data] if the local police departments are giving them the statistics,” Berk said. “You could say [for example] the Philadelphia police department is in cahoots with the Biden administration’s FBI, but that’s simply a silly conspiracy theory.”
Editor’s note: FactCheck.org does not accept advertising. We rely on grants and individual donations from people like you. Please consider a donation. Credit card donations may be made through our “Donate” page. If you prefer to give by check, send to: FactCheck.org, Annenberg Public Policy Center, 202 S. 36th St., Philadelphia, PA 19104.