FBI: 2025 Violent Crime Fell 9.3 Percent

The FBI’s preliminary 2025 national estimates showed violent crime down 9.3 percent, including an 18.1 percent decline in murder and non-negligent manslaughter.

Data Report Overview

The FBI’s Uniform Crime Reporting Program released its unusually early look at 2025 crime data on May 13, 2026. The release was important not because it announced one criminal case, but because it offered a national estimate based on data from more than 17,000 law-enforcement agencies covering about 96% of the U.S. population. More than 15,000 agencies submitted data through the National Incident-Based Reporting System, covering nearly 90% of the population.

The headline result was a broad decline. The FBI estimated that violent crime fell 9.3% from 2024 to 2025. Within that category, murder and non-negligent manslaughter fell 18.1%, rape fell 7.6%, robbery fell 18.5%, and aggravated assault fell 7.2%. Property crime fell an estimated 12.4%.

What the FBI Actually Reported

The strongest claim supported by the primary source is that the preliminary 2025 data show major declines across violent and property crime. The FBI also described the estimated one-year drops in violent crime and murder as the largest since 1937. That statement comes from the agency’s own historical comparison and should be attributed to the FBI rather than presented as an independently proven causal conclusion.

The FBI did not say its May release was the final 2025 annual crime report. It explicitly called the numbers preliminary and subject to change. That caveat matters because late agency submissions, revisions, estimation procedures and annual quality checks can alter the final rates. The early release is designed to provide timelier information without replacing the later full report.

Why the Murder-Rate Headline Is So Striking

Several analysts concluded that the preliminary 2025 decline likely pushed the national murder rate below the prior modern low in the FBI’s comparable series. Public-radio reporting quoted crime-data analyst Jeff Asher saying 2025 almost certainly produced the lowest murder rate recorded in the FBI-era data, which extend back to roughly the middle of the twentieth century with the most reliable modern comparisons commonly made from 1960 forward.

The claim should still be worded carefully. Saying the 2025 murder rate was likely the lowest in the comparable modern FBI record is stronger than saying the United States had the lowest homicide rate in all of its history. Older public-health records use different definitions, coverage and methodologies. PolitiFact noted that comparisons extending back 125 years are uncertain precisely because pre-1960 data are not directly comparable to the later FBI series.

Preliminary Does Not Mean Unreliable

Preliminary statistics are not guesses pulled from a small handful of cities. The FBI said more than 17,000 agencies contributed 2025 data and that those agencies covered 96% of the U.S. population. That is broad national coverage. The preliminary label reflects the timing of the release and the possibility of revisions, not an absence of underlying submissions.

At the same time, coverage percentages do not erase every measurement problem. Local agencies differ in reporting systems, offense classification, completeness and revision schedules. National estimates also summarize enormous geographic variation. A nationwide decline can coexist with increases in individual cities or categories, and a local rise does not disprove a national fall.

The FBI Crime Data Explorer Adds a Rolling View

The FBI’s Crime Data Explorer provides more current rolling trend information in addition to the calendar-year preliminary report. As of July 15, 2026, the CDE showed reported U.S. violent-crime trends for April 2025 through March 2026, including an 18.2% decline in murder, an 8.2% decline in rape, an 18.8% decline in robbery and a 6.8% decline in aggravated assault. Property-crime categories were also down in that rolling comparison.

Those numbers are useful, but they are not the same measurement window as the 2025 annual estimates. The May FBI release compares calendar 2025 with calendar 2024. The July CDE snapshot compares a rolling twelve-month period ending in March 2026 with the preceding period. Mixing the two would create a false precision. They point in the same broad direction, but they answer slightly different questions.

What the Submitted Analysis Argued

The submitted July article, a republished analysis by University of Montana criminologist James Tuttle, argued that the murder decline followed falling alcohol consumption, easing drug abuse and lower firearm purchasing after those factors rose during the pandemic-era violence spike. That is an interpretation of why homicide changed, not a causal finding announced by the FBI.

Tuttle’s argument is useful because it gives readers a theory to evaluate. He noted that homicide, substance abuse and gun purchasing rose during the period around 2020 and later fell. He also argued that lower youth criminal involvement may support continued declines. The article connects several national trends, but observational movement in the same direction cannot by itself prove that any one factor caused the murder-rate change.

Other Explanations Are Still Being Debated

Crime trends rarely have a single switch. Researchers have examined policing practices, incarceration patterns, violence-interruption programs, emergency medical care, demographic change, economic conditions, school attendance, drug markets, gun carrying, social disruption and the normalization that followed the acute pandemic period. The contribution of each factor can differ by city and time period.

That is why a national data article should separate measurement from explanation. The FBI can estimate how much reported crime changed. Determining why it changed requires additional research, natural experiments, local comparisons and years of evidence. A political official, police leader, advocacy group or academic may offer a theory, but that theory should not be written as though it were the FBI’s statistical conclusion.

2020 and 2021 Make the Decline Easier to Misread

The United States experienced a sharp homicide increase in 2020 and remained at an elevated level in 2021. When murder began falling afterward, one possible interpretation was simple reversion from an abnormal spike. But the significance of 2025 is that the decline appears to have carried the national rate below the pre-pandemic baseline rather than merely returning to it.

That is why analysts are treating the current pattern as more than a temporary bounce. If the final 2025 rate lands below the previous modern low and 2026 continues downward, the national trajectory would represent both a reversal of the pandemic-era increase and a continuation of the longer decline in violence that began decades earlier.

What 2026 Data Suggest So Far

Public-radio reporting in late June said early 2026 indicators suggested murder was continuing to fall. Asher’s independently collected data from hundreds of agencies showed another large decrease during the first part of 2026. The FBI’s rolling CDE view through March also showed murder substantially lower than in the comparison period.

Those early 2026 figures should not be used to rewrite 2025 as final. They are a separate developing story. Summer months can materially affect homicide totals, agencies continue reporting on different schedules, and final annual estimates take time. The defensible conclusion as of August 6 is that both the preliminary 2025 report and early 2026 indicators point downward.

What the Data Do Not Say

National crime estimates do not tell readers that every neighborhood is safer, that every type of violence fell by the same amount, or that a particular policy produced the decline. They do not resolve whether reported crime and victimization moved identically. They also do not make individual crimes less serious simply because the national rate is lower.

A falling murder rate can coexist with communities experiencing concentrated violence. Conversely, a high-profile local shooting does not negate a national decline. Good crime reporting has to hold both truths at once: individual cases matter intensely, and population-level trends require much larger denominators than the week’s most visible headlines.

Why Source Definitions Matter

The FBI’s murder and non-negligent manslaughter category is a law-enforcement reporting measure. Public-health homicide data maintained by the Centers for Disease Control and Prevention use a different system based on death certificates and can produce somewhat different totals or historical comparisons. News stories sometimes use murder and homicide interchangeably even when the underlying datasets do not.

For this reason, historical claims should name the source series. Saying the preliminary FBI murder rate appears to be the lowest in the comparable modern FBI record is precise. Saying the United States has never had a lower homicide rate in any historical dataset is much harder to establish and goes beyond the evidence reviewed here.

Current Outcome

The FBI’s preliminary 2025 national estimates showed violent crime down 9.3 percent, including an 18.1 percent decline in murder and non-negligent manslaughter.