The Homicide Rate Has Dropped to Historic Lows. Contrary to What Politicians Say, No One Really Knows Why.

A new report from the Council on Criminal Justice (CCJ) suggests the national homicide rate this year will hit a historic low, continuing a downward trend that began in 2022. In January, the CCJ estimated that the 2025 homicide rate would be the lowest since at least 1900, and the latest numbers indicate that the 2026 rate will be even lower. Other kinds of crime are also down, in some cases dramatically.

That development is surely welcome news. But we should be skeptical of the explanations offered by politicians of both major parties, who habitually reduce a complex reality to evidence in favor of the policies they have long supported.

The CCJ collects crime data from 36 U.S. cities, ranging in size from Rochester (population: 200,000) to New York City (8.5 million). The homicide estimates are based on data from 30 cities that have consistently reported numbers for that crime since 2018.

After rising slightly in 2019, the average homicide rate in the CCJ sample shot up in 2020, consistent with FBI data indicating a 30 percent increase that year. The rate in the CCJ sample rose again in 2021 and fell slightly in 2022 before dropping dramatically in 2023, 2024, and 2025. According to the new report, the average homicide rate for the first half of this year was down 18 percent from the same period in 2025 and down 51 percent since the first half of 2022.

This trend goes beyond a reversion to the pre-2020 norm. “Across the sample cities,” the CCJ says, “the 2026 average homicide rate” for January through June was “about 31% lower than in 2019.”

The FBI’s national numbers tell a similar story: Homicides fell by 11.6 percent in 2023, nearly 15 percent in 2024, and 18.1 percent in 2025 (based on a preliminary estimate). According to the FBI’s preliminary data, murders during the year ending in April 2026 were down 18.2 percent from the previous 12 months.

A longer view puts these numbers in perspective. Between 1991 and 2014, according to the CCJ’s national numbers, the homicide rate fell by more than 50 percent, then rose for a few years. But as of 2019, it was still about half the 1991 peak. The estimated rate for 2025, 4 per 100,000 residents, is even lower than the 2014 number (4.4), and it seems this year’s rate will be lower still.

Looking further back, the lowest homicide rate recorded since 1900, according to the CCJ’s graph, was 4.5 per 100,000, the rate in 1957 and 1958. Those numbers come from criminologist Randolph Roth’s 2009 book American Homicide. “These data rely primarily on death registration records,” the CCJ says, “and are widely regarded as the most reliable source for estimating homicide trends prior to the modern era of national crime reporting.”

A Bureau of Justice Statistics report lists somewhat lower rates for the 1950s, as low as 4.0 in 1957. But “when the FBI went back with a new methodology in 1958,” crime analyst Jeff Asher notes, “they looked at data from 1957 again and estimated that the actual 1957 murder rate was at 4.7 per 100,000.” In other words, it looks like the current homicide rate is even lower than the historic lows recorded during the 1950s.

The CCJ also reports large declines in three other crimes based on data for the first half of each year. The carjacking and car theft rates, both of which peaked in 2023, have since fallen by 73 percent and 51 percent, respectively. The residential burglary rate in the first half of 2026 was 56 percent lower than its 2018 peak.

Comparing the first half of 2026 to the first half of 2025, the CCJ reports declines in aggravated assault (2 percent), gun assault (6 percent), larceny (6 percent), nonresidential burglary (10 percent), residential burglary (13 percent), robbery (17 percent), car theft (20 percent), and carjacking (47 percent). By contrast, sexual assault, shoplifting, and domestic violence rose by 3 percent, 4 percent, and 8 percent, respectively.

Overall, violent crime from January through June was “at or below levels in the first half of 2019,” the CCJ says. Nonviolent crime also was “generally below levels seen in the first six months of 2019.”

President Donald Trump would like to take credit for these positive trends. “Last year, the murder rate saw its single largest decline in recorded history,” he bragged during his 2026 State of the Union address. “This is the biggest decline, think of it, in recorded history. The lowest number in over 125 years.”

Inconveniently for Trump, the 2020 murder surge happened on his watch, and the rate began falling before he started his second term. In any case, it is doubtful that presidential policies have much impact on crime rates. Although Trump credits immigration enforcement, for example, his deportation campaign, contrary to the way he portrays it, is by no means carefully targeted at unauthorized residents who pose a threat to public safety.

The evidence also casts doubt on policy prescriptions that Democrats tend to offer. “Private gun ownership in America has surged in the past several years,” the CCJ notes. “Meanwhile, violent crime, particularly homicide, saw a historic rise followed by a historic decline.”

The CCJ suggests a few more plausible theories, although none is completely satisfying. “A range of crime researchers and practitioners broadly agree that there’s no single explanation for the decline” in homicides, it says. “Instead, it reflects multiple forces moving in the same direction, from prevention efforts and law enforcement strategies to broader social changes following the pandemic shock.”

Last February in USA Today, CCJ President Adam Gelb cautioned against “partisan platitudes” that offer simplistic explanations for the ongoing crime drop. “Murder and other crime are falling in cities that are governed by different political philosophies, operating under varied policing and prosecution models and facing distinct local economic conditions,” he noted. “Local policies and programs surely matter. But the striking consistency of the crime decline across the country suggests that macro-level forces are exerting enormous influence as well.”

It is plausible that the 2020 homicide spike, like the concomitant surge in drug-related deaths, had something to do with the social, economic, and educational upheaval caused by the COVID-19 pandemic and the government’s response to it. “As schools, workplaces, social programs, churches and civic institutions regained footing after COVID-19 disruption,” Gelb noted, “emotional and economic stresses eased and daily routines strengthened.”

Gelb added that “courts slowed by COVID-19 disruptions have largely resumed normal function”; that “police departments, on their heels after the protests of 2020, have reemphasized targeted enforcement strategies”; and that “communities have scaled up violence prevention efforts and prisoner reentry programs to unprecedented levels, aided in part by federal recovery funds.” He conceded that “immigration enforcement has removed dangerous offenders” but added that it has swept up “many people who pose little or no public safety risk.”

Gelb also credited “the spread of public and private surveillance cameras and license plate readers,” which he said had “deterred would-be offenders while helping police solve more cases.” He added that “the transition to digital transactions means fewer people carry cash, making them less attractive targets of a robbery.” And he suggested that “the gravitational pull of cyberspace” may be “altering social routines historically associated with street violence.”

Gelb mentioned several other possible factors, including changes in illicit drug markets, reduced alcohol consumption, less exposure to “environmental toxins” such as lead, and the “grim demographic reality” that homicides and drug overdoses may have “removed tens of thousands of high-risk people from the population.” But he noted that “researchers are still struggling to confidently explain the long-term crime drop that began in the early 1990s,” which suggests that “definitive answers about current trends will take time.”

The honest answer to the question of why crime is falling, in short, is that we don’t really know and may never be sure. As tempting as it may be for politicians, it is not safe to assume that any particular government policies have played an important role in the absence of evidence that they actually work.