Skip to content

Crime · United States · Updated 2026-10-26

Is crime rising or falling in the United States?

The FBI's police-recorded violent crime rate fell 14.3% from 2015 to 2025, while the victim survey's rate of violent crime excluding simple assault rose 30.9% from 2015 to 2024; property crime fell 37.9% in FBI records and fell 11.8% in the survey over the same spans.

Every figure in this summary is calculated from the sources listed on this page.

Violent crime, change since 2015

Change in the rate of violent crime (rape, robbery and aggravated assault; the FBI also counts murder), measured from police records (FBI) and from a victim survey that counts crimes whether or not they were reported (NCVS). United States, 2015 to 2025 for the FBI; 2015 to 2024 for the victim survey, whose redesigned 2025 estimates are not comparable. Unit: % change in the rate.

Sources disagree% change in the rate, 2025

73Trust scoreModerate certainty

Score breakdown

Quality32 / 40

How well the sources are made

Agreement15 / 30

How close the sources are to each other

Independence16 / 20

How many separate ways the number was measured

Freshness10 / 10

How recent the newest data is

How much each source counts

  • FBI50%
  • NCVS50%

Sources

SourceValueDefinitionQualityWeight
FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1)Federal Bureau of Investigation · Police and justice recordsRetrieved 2026-10-26Falling −14.3%2025
Estimated violent crime offenses known to law enforcement (murder and nonnegligent manslaughter, rape under the revised definition, robbery, aggravated assault) per 100,000 residents: 382.17 in 2015 (1,228,353 offenses) and 327.62 in 2025 (1,119,768).

Rates computed from the counts and populations in CIUS Table 1, 2006-2025 (Crime Data Explorer, Documents & Downloads, Crime in the United States 2025, CIUS Estimations).

The chart leaves out 2021 because the FBI says that year is not comparable with others. Murder was 1.3% of the violent crimes in 2025.

4 of 5 checks
  • Yes: Method and definitions published. The FBI publishes its estimation method (CIUS Methodology, in the same download) and offense definitions.
  • Yes: Official statistics or peer-reviewed. Official national crime statistics of the FBI's Uniform Crime Reporting Program.
  • Yes: Full count or representative sample. Built from offenses recorded by law enforcement agencies covering most of the population, with FBI estimates for agencies that did not report; it counts only crimes known to police.
  • Yes: Raw data downloadable. National, state and agency tables are downloadable from the Crime Data Explorer.
  • No: No known break or quality problem. The move to NIBRS left 2021 short of data: reporting agencies covered 74.12% to 79.51% of the population each month (at least 93.65% in every other month of 2015-2025), and the FBI says its 2021 estimates are not nationally representative or comparable with other years (CIUS Table 1, note 4). Earlier years are re-estimated when agencies send late data.
50%
BJS National Crime Victimization Survey (legacy questionnaire): violent crime excluding simple assaultBureau of Justice Statistics · Victim surveyRetrieved 2026-10-26Rising +30.9%2024
Violent victimizations excluding simple assault (rape or sexual assault, robbery, aggravated assault) per 1,000 persons age 12 or older, reported to police or not: 6.8 in 2015 (95% CI 5.61-7.95) and 8.9 in 2024 (7.1-10.67).

Simple assault is left out so the offenses match the FBI's violent crime list, which has no simple assault; murder cannot be measured by a victim survey. Rates as published in N-DASH; 2020-2024 match Table 1 of Criminal Victimization, 2024.

Including simple assault, the violent victimization rate went from 18.6 to 23.3 per 1,000 (+25.3%). 2024 is the last year of the legacy survey.

4 of 5 checks
  • Yes: Method and definitions published. BJS publishes the survey method, definitions and standard errors (Criminal Victimization, 2024, Methodology).
  • Yes: Official statistics or peer-reviewed. Official statistics from the Bureau of Justice Statistics; interviews are done by the U.S. Census Bureau.
  • Yes: Full count or representative sample. Nationally representative sample of households, interviewing everyone age 12 or older: in 2024 the legacy half-sample had 69,396 household interviews (60% of eligible households) and 109,341 person interviews.
  • Yes: Raw data downloadable. Yearly estimates with confidence intervals download as CSV from the NCVS Dashboard (N-DASH); microdata are archived at NACJD.
  • No: No known break or quality problem. BJS had to revise the 2016 data after a sample expansion made comparison with earlier years difficult; in 2024 only about half the sample got the legacy questionnaire, which widened the error margins; the series stops in 2024 because the redesigned survey used from 2025 is not comparable.
50%
Violent crime, change since 2015, by source · rate index, 2015 = 100
  • FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1)
  • BJS National Crime Victimization Survey (legacy questionnaire): violent crime excluding simple assault
050100150201520172019202120232025FBIFBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2015: 100FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2016: 103.8FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2017: 101.1FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2018: 98FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2019: 96.4FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2020: 101.8FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2022: 101.9FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2023: 99.9FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2024: 95FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2025: 85.7NCVSBJS National Crime Victimization Survey (legacy questionnaire): violent crime excluding simple assault, 2015: 100BJS National Crime Victimization Survey (legacy questionnaire): violent crime excluding simple assault, 2016: 97.1BJS National Crime Victimization Survey (legacy questionnaire): violent crime excluding simple assault, 2017: 107.4BJS National Crime Victimization Survey (legacy questionnaire): violent crime excluding simple assault, 2018: 126.5BJS National Crime Victimization Survey (legacy questionnaire): violent crime excluding simple assault, 2019: 107.4BJS National Crime Victimization Survey (legacy questionnaire): violent crime excluding simple assault, 2020: 82.4BJS National Crime Victimization Survey (legacy questionnaire): violent crime excluding simple assault, 2021: 82.4BJS National Crime Victimization Survey (legacy questionnaire): violent crime excluding simple assault, 2022: 144.1BJS National Crime Victimization Survey (legacy questionnaire): violent crime excluding simple assault, 2023: 127.9BJS National Crime Victimization Survey (legacy questionnaire): violent crime excluding simple assault, 2024: 130.9
Show the data as a table
YearFBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1)BJS National Crime Victimization Survey (legacy questionnaire): violent crime excluding simple assault
2015100100
2016103.897.1
2017101.1107.4
201898126.5
201996.4107.4
2020101.882.4
2021–82.4
2022101.9144.1
202399.9127.9
202495130.9
202585.7–

Property crime, change since 2015

Change in the rate of property crime (burglary, motor vehicle theft and other theft), measured from police records (FBI) and from a victim survey that counts crimes whether or not they were reported (NCVS). United States, 2015 to 2025 for the FBI; 2015 to 2024 for the victim survey, whose redesigned 2025 estimates are not comparable. Unit: % change in the rate.

Falling% change in the rate, 2025

88Trust scoreHigh certainty

Score breakdown

Quality32 / 40

How well the sources are made

Agreement30 / 30

How close the sources are to each other

Independence16 / 20

How many separate ways the number was measured

Freshness10 / 10

How recent the newest data is

How much each source counts

  • FBI50%
  • NCVS50%

Sources

SourceValueDefinitionQualityWeight
FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1)Federal Bureau of Investigation · Police and justice recordsRetrieved 2026-10-26Falling −37.9%2025
Estimated property crime offenses known to law enforcement (burglary, larceny-theft, motor vehicle theft; arson is not estimated) per 100,000 residents: 2469.81 in 2015 (7,938,449 offenses) and 1534.82 in 2025 (5,245,768).

Rates computed from the counts and populations in CIUS Table 1, 2006-2025 (Crime Data Explorer, Documents & Downloads, Crime in the United States 2025, CIUS Estimations).

The chart leaves out 2021 because the FBI says that year is not comparable with others.

4 of 5 checks
  • Yes: Method and definitions published. The FBI publishes its estimation method (CIUS Methodology, in the same download) and offense definitions.
  • Yes: Official statistics or peer-reviewed. Official national crime statistics of the FBI's Uniform Crime Reporting Program.
  • Yes: Full count or representative sample. Built from offenses recorded by law enforcement agencies covering most of the population, with FBI estimates for agencies that did not report; it counts only crimes known to police.
  • Yes: Raw data downloadable. National, state and agency tables are downloadable from the Crime Data Explorer.
  • No: No known break or quality problem. The move to NIBRS left 2021 short of data: reporting agencies covered 74.12% to 79.51% of the population each month (at least 93.65% in every other month of 2015-2025), and the FBI says its 2021 estimates are not nationally representative or comparable with other years (CIUS Table 1, note 4). Earlier years are re-estimated when agencies send late data.
50%
BJS National Crime Victimization Survey (legacy questionnaire): property victimizationBureau of Justice Statistics · Victim surveyRetrieved 2026-10-26Falling −11.8%2024
Household property victimizations (burglary or trespassing, motor vehicle theft, other theft) per 1,000 households, reported to police or not: 110.7 in 2015 (95% CI 105.54-115.91) and 97.6 in 2024 (92.76-102.47).

Rates as published in N-DASH; 2020-2024 match Table 2 of Criminal Victimization, 2024.

2024 is the last year of the legacy survey.

4 of 5 checks
  • Yes: Method and definitions published. BJS publishes the survey method, definitions and standard errors (Criminal Victimization, 2024, Methodology).
  • Yes: Official statistics or peer-reviewed. Official statistics from the Bureau of Justice Statistics; interviews are done by the U.S. Census Bureau.
  • Yes: Full count or representative sample. Nationally representative sample of households: in 2024 the legacy half-sample had 69,396 household interviews (60% of eligible households) and 109,341 person interviews.
  • Yes: Raw data downloadable. Yearly estimates with confidence intervals download as CSV from the NCVS Dashboard (N-DASH); microdata are archived at NACJD.
  • No: No known break or quality problem. BJS had to revise the 2016 data after a sample expansion made comparison with earlier years difficult; in 2024 only about half the sample got the legacy questionnaire, which widened the error margins; the series stops in 2024 because the redesigned survey used from 2025 is not comparable.
50%
Property crime, change since 2015, by source · rate index, 2015 = 100
  • FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1)
  • BJS National Crime Victimization Survey (legacy questionnaire): property victimization
050100150201520172019202120232025FBIFBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2015: 100FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2016: 98.9FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2017: 94.7FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2018: 88.5FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2019: 84.3FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2020: 78.1FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2022: 79.8FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2023: 78.5FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2024: 71.3FBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1), 2025: 62.1NCVSBJS National Crime Victimization Survey (legacy questionnaire): property victimization, 2015: 100BJS National Crime Victimization Survey (legacy questionnaire): property victimization, 2016: 107.1BJS National Crime Victimization Survey (legacy questionnaire): property victimization, 2017: 97.9BJS National Crime Victimization Survey (legacy questionnaire): property victimization, 2018: 97.7BJS National Crime Victimization Survey (legacy questionnaire): property victimization, 2019: 91.6BJS National Crime Victimization Survey (legacy questionnaire): property victimization, 2020: 85.4BJS National Crime Victimization Survey (legacy questionnaire): property victimization, 2021: 81.6BJS National Crime Victimization Survey (legacy questionnaire): property victimization, 2022: 92.1BJS National Crime Victimization Survey (legacy questionnaire): property victimization, 2023: 92.3BJS National Crime Victimization Survey (legacy questionnaire): property victimization, 2024: 88.2
Show the data as a table
YearFBI Uniform Crime Reporting Program, national estimates (Crime in the United States, Table 1)BJS National Crime Victimization Survey (legacy questionnaire): property victimization
2015100100
201698.9107.1
201794.797.9
201888.597.7
201984.391.6
202078.185.4
2021–81.6
202279.892.1
202378.592.3
202471.388.2
202562.1–

Why the numbers differ

  • Many crimes never reach police records. In the 2024 survey, victims said 47.9% of violent victimizations, 60.2% of violent crimes excluding simple assault and 30.5% of property victimizations were reported to police (BJS, Criminal Victimization, 2024, Table 4). For violent crime excluding simple assault the share was 50.6% a year earlier, so changes in reporting can move police figures on their own. (source, retrieved 2026-10-26)
  • FBI coverage dropped in 2021, when its estimates rest mainly on NIBRS submissions: agencies reporting to it covered 74.12% to 79.51% of the population each month, against at least 93.65% in every other month from 2015 to 2025. The FBI says its 2021 national estimates are not nationally representative or comparable with other years. (source, retrieved 2026-10-26)
  • The survey interviews people age 12 or older living in households, so it cannot count murders (1.3% of the violent crimes in FBI figures for 2025) and leaves out people in prisons, hospitals and other institutions, people on military bases and children under 12 (BJS, Criminal Victimization, 2024, Methodology, p. 19). (source, retrieved 2026-10-26)
  • Survey estimates carry sampling error. The 95% confidence interval for violent crime excluding simple assault was 5.61-7.95 per 1,000 persons in 2015 and 7.1-10.67 in 2024; the two intervals overlap. For property victimization they were 105.54-115.91 and 92.76-102.47 per 1,000 households and do not overlap. In 2024 only about half the sample answered the legacy questionnaire, which widened the intervals. (source, retrieved 2026-10-26)
  • The sources end in different years. Over the survey's span, 2015 to 2024, the FBI violent crime rate changed -5.0% and the FBI property crime rate -28.7% (CIUS Table 1). (source, retrieved 2026-10-26)
  • The survey counts crimes by interview year, not the year they happened: the 2024 survey covers crimes from July 2023 to November 2024 (BJS, Criminal Victimization, 2024, Methodology, p. 19). (source, retrieved 2026-10-26)

Common mistakes

  • Comparing the two levels. The FBI rate is per 100,000 residents of all ages and counts only crimes known to police; the survey rate is per 1,000 persons age 12 or older (violent) or per 1,000 households (property) and includes crimes never reported. Compare their directions, not their sizes. (source, retrieved 2026-10-26)
  • Joining the redesigned survey to the old series. BJS moved fully to a new questionnaire (NCVS-R) in January 2025 and says NCVS-R estimates are not comparable with legacy NCVS estimates. (source, retrieved 2026-10-26)