On multiple occasions throughout 2024, U.S. President Joe Biden and members of his administration claimed that the violent crime rate in the United States was at a 50-year low. Biden even included the statistic in his 2024 State of the Union Address in March:
The year before I took office, murders went up 30% nationwide the biggest increase in history.
That was then.
Now, through my American Rescue Plan, which every Republican voted against, I've made the largest investment in public safety ever.
Last year, the murder rate saw the sharpest decrease in history, and violent crime fell to one of the lowest levels in more than 50 years.
But as the Biden administration continued to advertise the statistic, former President Donald Trump pushed back against it, claiming it was an outright lie, and that U.S. crime rates were rising, not falling.
Trump's followers took up his claim and promoted it. When White House Press Secretary Karine Jean-Pierre talked about record-low crime rates in a news conference on Aug. 14, many on social media left comments stating that the statistic was wrong, alleging that the Biden administration was failing to accurately record crime and therefore incorrect about the true crime rate.
The Problem
Collecting accurate statistics on crime is hard, because by definition, committing a crime is illegal. Perpetrators don't regularly admit to their illegal activities, nor are all crimes reported to law enforcement authorities. As such, there are various sources that report crime statistics across the U.S., each of which has its own shortcomings.
The Biden administration's main source for crime statistics was the FBI, which has been collecting data on crime for over 100 years through its Uniform Crime Reporting (UCR) program. The program records statistics for eight different crimes, divided into two categories: Murder and homicide, rape, robbery and aggravated assault are classified as "violent crimes," while arson, burglary, larceny and motor vehicle theft are classified as "property crimes." The data on these crimes dating back to 1979 is publicly available through the FBI's Crime Data Explorer.
Said data strongly suggests that Biden's claim of a 50-year low in violent crime is at least roughly accurate. However, there are notable limitations to those statistics that prevent us from drawing a definitive conclusion.
First, the FBI broadened its definition of rape in 2013, making it difficult to accurately compare pre-2013 and post-2013 data in that category.
Second, the way the FBI collects its data yields results that undercount the "true" figure of crime.
In order for an incident to be included in FBI data, the crime must be reported to a police department that participates in the UCR program. Since crime is not always reported to law enforcement (in fact, most crime goes unreported), the FBI's numbers will never be 100% complete and accurate using that methodology.
Furthermore, law enforcement agencies aren't required to report their data to the FBI, which many commenters on social media used as a justification for claiming that if the crime rate "fell," it must have been because police departments stopped reporting their data. However, the FBI also publishes the list of law enforcement agencies participating in the UCR program. According to Pew Research, 83% of law enforcement agencies sent data to the FBI in 2022.
Regardless, the fact that not all agencies report to the FBI means that the agency must extrapolate from the data it does receive in order to create its estimates, and any kind of extrapolation creates additional uncertainty in the numbers.
Finally, and most importantly for the claim at hand, the FBI's definition of "violent crime" is quite limited, consisting of only murder and manslaughter, rape, robbery (theft with a weapon) and aggravated assault (assault with a weapon).
Alternative data sources often add additional charges to their definitions of "violent crime," like simple assault and intimidation. Differences in the way violent crime is defined make it harder to accurately compare the FBI's data with data from other sources.
What the FBI's Data Says
Because the Biden administration exclusively used FBI statistics to support its claim, we created visualizations using the same data (see below). The FBI calculates the crime rate based on the number of reported crimes for every 100,000 people. (We should note that no
Based on the original 2022 FBI data, violent crime fell in 2022 as well. However, the agency updated its 2022 statistics when it released 2023 data, a fact publicized in October 2024 by conservative media outlets like Fox News and the New York Post. Those updated numbers showed that violent crime rose in 2022 before falling in 2023. This sudden change is not particularly surprising. The FBI's methodology, which records only reported crimes, is and always will be vulnerable to under-counting.
But even with more recent data, the Biden administration's claim of a 50-year low is likely accurate.
We originally calculated that in 2022, there were 380.7 violent crimes for every 100,000 people. Using the FBI's updated numbers, we found that the rate jumped to 388.7 violent crimes for 100,000 people in 2022. In 2023, however, we calculated a violent crime rate of 374.4. That rate was marginally lower in 2014 (372), meaning that at best, we can conclude that the crime rate was "near" a 50-year low. Although it's not necessarily proper to compare two different data sets, PolitiFact calculated that the violent crime rate was 401.1 in 1972 and 461.1 in 1974, and it did not fall below 400 until 2011.
Furthermore, the FBI has periodically announced preliminary statistics on crime data for 2023 and 2024 through news releases. We decided to not include those preliminary statistics from 2023 and 2024 i
If the FBI's preliminary figures are accurate, however, that decrease would likely verify a 50-year low
In order to look at the data in more detail, we also produced an interactive map showing how the violent crime rate has differed between states since 1979.
Almost all states follow the general pattern of a peak in the 1980s or 1990s and a gradual decrease since, as shown by the lighter colors across the country. Even in states where crime has been trending upward in recent years, like New Mexico, they still had higher peaks in the 1990s. The only state where 2022's violent crime rate was the highest on record was Vermont, with 221.9 violent crimes per 100,000 people. That rate decreased to 210.4 in 2023.
As a comparison,
The other main crime data source, the National Crime Victimization Survey (NCVS), began in the 1990s and largely corroborates FBI data. A project of the Bureau of Justice Statistics (BJS), the NCVS attempts to track both reported and unreported crimes by conducting interviews with a large, national sample of people ages 12 and over. This is a key distinction between the two data sources.
Individuals interviewed for the NCVS are asked whether they were the victim of a crime in the last six months, with follow-up questions collecting details about the crime in question and whether or not it was reported to the police. Because the NCVS also investigates unreported crime, some sources consider it to be more accurate or a better source of information for national crime statistics. However, it too has limitations.
Since the NCVS is a survey, researchers have to extrapolate from its data in order to create estimates for national and state statistics. Just like the FBI, this creates a layer of uncertainty around the survey numbers. Second, the BJS does not interview enough people in each state to create accurate estimates at a state-wide level. Finally, because the BJS is only able to ask people if they were affected by a crime after that crime has been committed, it does not record or collect any statistics about murder and homicide. This means that FBI and NCVS numbers cannot be directly compared.
However, we are able to compare the long term trends in crime between the two data sets. Since the NCVS records more specific information about each crime that occurred, Snopes attempted to factor in only crimes that would be considered "violent crimes" by both the FBI and the NCVS in order to create the following graphic. We used data from 1992, when the NCVS started, to its most recent published data, 2023:
Even though the NCVS data only goes as far back as 1992, the general trend over time since 1992 is roughly the same: Violent crime has decreased since its peak in the 1990s.
John Lott, the founder of a right-wing gun-advocacy group called the Crime Prevention Research Center, has used this data to criticize the Biden administration's crime statistics in op-eds. Both Lott and Trump specifically pointed to the 2022 NCVS, the most recent data available at the time, as alleged proof that the FBI's statistics are untrustworthy, and that violent crime was actually rising in 2024.
However, the assertion that crime is rising is based only on the 2022 data point. Since the data necessary to conclude violent crime had risen across the country since then had not yet been released, these claims, including those made by Trump, were unfounded. Furthermore, according to 2023 NCVS data, the violent crime rate remained roughly the same from 2022 to 2023.
It's worth noting that the significant jump from 2021 to 2022 left even researchers confused. The Marshall Project,
An immediate explanation is quite simple: The 2022 NCVS found that fewer crimes were reported to the police overall. Academics have proposed several theories as to why people reported fewer crimes, but researchers are waiting on more data before attempting to draw conclusions.
Over a long period of time, the data shows that as of 2022-23, reported violent crime consistently decreased since it peaked in the 1990s. On a shorter time scale, however, we cannot draw a strong conclusion.
Based on the FBI data, the Biden administration's claim of a 50-year low in violent crime is likely correct. The overall conclusion of our analysis has not significantly changed since the FBI released its 2023 data — although that updated data shows a slight rise in violent crime rate in 2022, the administration's claim is about violent crime in 2024. However, the claim comes with limitations, since the FBI tracks only reported crime.
Meanwhile, Trump's claim that violent crime was rising in 2024 was unfounded, because the data he would need to make such a claim did not exist yet. While the 2022 NCVS data did show a jump in violent crime, it's irresponsible to draw a large conclusion off of one data point. In a news release published alongside the 2023 NCVS data, BJS director Kevin M. Scott said that the 2023 rate was no different statistically from where it was in 2019, before the COVID-19 pandemic.