Policing: By the Numbers
Updated May 2026
Originally Published February 2021
Over the past dozen years, American policing has been the subject of sustained public scrutiny and debate. High-profile incidents involving the use of force by law enforcement, as well as longstanding concerns about accountability and equity, have intensified calls for reform, changes in funding priorities, and expanded investments in community-based safety and social services. At the same time, many U.S. cities experienced sharp increases in violent crime beginning in 2020, the first year of the COVID-19 pandemic, followed more recently by significant declines, underscoring the challenge of advancing both public safety and justice.
Efforts to develop effective responses that achieve the twin goals of crime control and justice must be grounded in reliable data and research evidence, as well as personal and professional experience. This series of charts informs discussions about the future of policing in America by providing a statistical portrait of key trends.
The charts examine changes in the size and composition of police agencies, spending on policing, reported crime and victimization rates, people killed by police, and officers killed in the line of duty, as well as shifts in public perceptions of and trust in law enforcement. Time periods vary due to data availability. Trends by race and ethnicity are presented where reliable data are available.
Taken together, the metrics packaged here illuminate broad patterns in policing over time as governments, agencies, and communities continue to wrestle with priorities, tradeoffs, and reform strategies at the local level. The overall picture is complex and nuanced, reflecting both the scale of the challenges facing policing nationwide and the diversity of conditions across communities.
Key Takeaways
- Police staffing has grown substantially over the long term but remains below recent peaks. The number of sworn officers roughly doubled between 1980 and its peak in 2017. Staffing dropped after 2017 and has not fully returned to pre-pandemic levels, though hiring has rebounded in recent years.
- Women remain underrepresented in policing despite steady gains. The share of female sworn officers has nearly tripled since 1980, rising from about 5% to 14% of officers, though the rate of increase has slowed.
- Relative to crime levels—but not population size—the U.S. had far more police officers in 2024 than in previous decades. While the number of officers increased over time, serious crime fell sharply after the 1990s. As a result, the number of officers relative to the volume of serious crime is more than 3.5 times higher today than in 1980.
- Police spending has increased significantly, but its share of government budgets has remained stable. After adjusting for inflation, state and local police spending roughly tripled between 1977 and 2023. However, police spending has consistently accounted for just over 3% of total state and local government expenditures.
- Police-public contacts declined in recent years, with notable shifts in who is most likely to experience them. Across multiple types of contact, including police-initiated stops, resident-initiated calls, and traffic accident responses, rates of interaction between police and the public were substantially lower in 2022 than in earlier years. Racial patterns also shifted: In 2015, Black residents experienced the highest rates of police-initiated contact, but from 2018 to 2022, White residents had the highest rates.
- Fatal police encounters have increased gradually over time, and patterns differ by race and ethnicity. White people account for the largest share of those killed by police (about 45% in a typical year), followed by Black (about 25%) and Hispanic individuals (about 18%). When adjusted for population size, Black and American Indian/Alaska Native people are killed at the highest rates.
- Fatal assaults on officers remain relatively rare, while nonfatal assaults have increased in recent years. Between 25 and 80 officers are fatally assaulted each year, with no sustained long-term increase. Officer assaults rose sharply after 2021, reaching about 86,000 in 2024.
- Public perceptions of police differ sharply by race. White Americans consistently report higher levels of confidence in the police and more favorable perceptions of fairness and respect than Black Americans. At the same time, Black Americans have generally been more likely to report violent crime to police.
Table of Contents
Tip for Exploring the Charts
Many figures display multiple groups at once (for example, Black and White residents, or several offense categories). To focus on a single group, you can click on any label in the chart’s legend—such as “White” or “Property Crime”—to show or hide that series. This feature lets you simplify the view, compare specific groups, or follow one trend line more easily.
Policy Agency Size and Composition
Number of Officers
The number of sworn police officers roughly doubled between 1980 and its peak in 2017, before dropping in 2018 and fluctuating through 2021 (Figure 1). Staffing began to increase again in 2022, continuing through 2024. The decline after 2017 likely reflects, in part, changes in reporting coverage rather than a sharp drop in staffing.1 At the same time, law enforcement agencies began reporting growing recruitment and retention challenges during this period, including fewer applicants, increases in retirements, and higher attrition.2 In 2024, there were about 770,000 sworn law enforcement officers—roughly 30,000 fewer than in 2017.
Growth in staffing did not keep pace with population expansion. The rate of officers per 100,000 residents peaked in 2009, largely plateaued through 2017, and then declined through 2021. The rate rose modestly in recent years. Even with those increases, there were about 11% fewer officers per capita in 2024 than in 2009.
At the same time, long-term declines in serious crime altered the relationship between police staffing and crime volume. Relative to serious crime—defined as the FBI’s Part I index offenses (homicide, robbery, rape, aggravated assault, burglary, larceny, motor vehicle theft, and arson)—the number of officers was 3.5 times higher in 2024 than in 1980. The rate of sworn officers per 1,000 Part I crimes was 107.0 in 2024, compared with 30.3 in 1980.
Figure 1. Full-Time, Sworn Law Enforcement Personnel, 1980-2024
More recent survey data from the Police Executive Research Forum (PERF), a national policing research and membership organization, provide additional insight into staffing trends. The survey is based on responses from agencies representing more than 128,000 sworn officers and indicates that staffing declined 5.5% between 2020 and 2024 and remained below 2020 levels in 2025, despite a small increase from 2024 to 2025 (Figure 2).
Figure 2. Percent Change in Police Staffing From 2020-2025
Although overall staffing has not fully returned to 2020 levels, PERF data show that hiring has rebounded (Figure 3). In 2024, agencies responding to PERF’s survey hired 12.5% more officers than in 2019 and 5% more than in 2023.
Figure 3. Police Hiring Trends, 2019-2024
Female Officers
Both the number and share of female sworn officers increased substantially from 1980 to 2017, with a brief spike in 2011 (Figure 4). After declining in 2018 and leveling off through 2021, the numbers began rising again in 2022 and continued increasing through 2024. Meanwhile, the share increased consistently over that same time. In total, the number of female officers grew from about 20,000 in 1980 to 111,000 in 2024, while their share of total officers nearly tripled from 5% to 14%. Most of these gains occurred during the 1990s; a long period of relative stability followed, before the more recent increases.
Figure 4. Share of Female Officers in Local Police Agencies, 1980-2024
Non-White Officers
The racial and ethnic composition of sworn officers differed in 2020 compared with 1987, though patterns varied across groups (Figure 5). National data on officer race and ethnicity are drawn from the Bureau of Justice Statistics’ Law Enforcement Management and Administrative Statistics survey, the only source that collects these data consistently across agencies; the most recent available year is 2020.
The share of Black officers was about 25% higher in 2020 than in 1987. In 2020, Black officers accounted for about 12% of sworn personnel, compared with 12% of the U.S. population. Changes in Hispanic representation were more pronounced. The share of Hispanic officers was 14% in 2020—more than three times their share in 1987. But that proportion remained below the share of Hispanic and Latino people in the general population (19%).
Data on officers identifying as American Indian or Alaska Native (AI/AN), Asian, Native Hawaiian, or Other Pacific Islander, or as two or more races were not collected consistently over time. However, people identifying with racial or ethnic groups other than Black, Hispanic, or White comprised less than 1% of sworn officers in 1987 and about 4% in 2020.
Comparisons between officer demographics and local population composition vary substantially by agency size and geography,3 highlighting the importance of examining police representation within local context rather than relying solely on national averages.
Figure 5. Share of Officers in Local Police Agencies by Race, 1987-2020
Police Spending
State and Local Expenditures
After adjusting for inflation to 2023 dollars, total state and local police expenditures roughly tripled between 1977 and 2023, rising from about $52.5 billion to $151.6 billion (Figure 6). Per capita spending nearly doubled over the same period, increasing from $241 per U.S. resident to $454. Despite this growth, police spending has consistently accounted for just over 3% of total state and local government expenditures.
Figure 6. State and Local Police Expenditures, 1977-2023
Police Contacts with the Public
Data on contacts between police and the public are drawn from a nationally representative survey of U.S. residents that asks about their most recent contact with police in the past 12 months. Because these data are based on self-reports, findings reflect how individuals describe and interpret their interactions with law enforcement.
Police-Initiated Contacts
National data on the nature and type of police-initiated contacts—and the reasons for those contacts—are not routinely collected through administrative systems. This survey helps fill that gap by asking respondents about encounters initiated by police. The most recent data reported by race and ethnicity are from 2022 (Figure 7).
In 2022, respondents most commonly reported that police-initiated contacts involved traffic stops. In absolute terms, White drivers and passengers accounted for substantially more reported traffic stops than Black or Hispanic people, reflecting their larger share of the population and driving exposure. When measured as rates, however, differences across groups were narrower: White respondents reported the highest traffic stop rate, followed closely by Black respondents and then Hispanic respondents.
A different pattern emerged for arrests. While more White respondents reported being arrested in absolute numbers, Black respondents reported the highest per capita arrest rate.
Figure 7. Police-Initiated Contacts by Race, 2022
Resident-Initiated Contacts
Residents also initiate contact with police, most commonly to report a possible crime (Figure 8). In 2022, survey respondents most frequently reported contacting police for suspected criminal activity, followed by contacts related to non-crime emergencies and other assistance.
In absolute terms, White respondents reported a substantially higher number of police contacts across all categories than Black or Hispanic respondents. These differences reflect, in part, population size and exposure.
When the data are examined as rates, however, differences across groups were more modest. White respondents reported higher rates of contact across most categories, including reporting possible crimes and non-crime emergencies. Black respondents reported the lowest rates of contact for these types of calls.
Patterns differed somewhat for other forms of resident-initiated contact. Rates of requests for non-emergency assistance or other help-seeking were more similar across groups. For block or neighborhood watch-related contact, rates were comparable overall, with Black respondents reporting a slightly higher rate than White respondents.
Figure 8. Resident-Initiated Contacts by Race, 2022
Police Contacts Over Time
Rates of contact between police and the public, as reported by respondents, were substantially lower in 2022 than in previous years across three contact types: police-initiated contacts, resident-initiated contacts, and contacts related to traffic accidents.
Traffic accident-related contacts are analyzed separately because they do not fall neatly into either police- or resident-initiated categories. Over this period, not only did overall rates decline, but the racial and ethnic distribution of reported contacts also shifted in important ways (Figure 9).
Police-Initiated Contacts
In 2015, Black respondents aged 16 and older reported the highest rates of police-initiated contact, followed closely by White respondents. Beginning in 2018, that pattern reversed, with White respondents reporting the highest rates. No clear pattern emerged for police-initiated contacts with Hispanic residents and those identifying with other races. In 2020, police-initiated contacts were the lowest for Hispanic residents, while in 2022, Hispanic residents were stopped at a slightly higher rate than Black residents. These changes occurred along with overall declines in police-initiated contact rates across all groups.
Resident-Initiated Contacts
Across all years examined, White respondents consistently reported the highest rates of resident-initiated contact with police. In 2015, Black respondents reported the second-highest rate of contact, but this pattern shifted gradually over time. By 2022, Black respondents reported the lowest rate of resident-initiated contact, though rates among Black and Hispanic respondents, as well as those identifying with other races, were closely clustered.
Traffic Accidents
Rates of police contact related to traffic accidents were more similar across racial and ethnic groups than were other forms of contact throughout the period. In 2022, Black people experienced the highest rate of traffic accident-related contact, though differences across groups were relatively modest.
Figure 9. Rate of Police-Public Contacts by Race, 2015-2022
Crime Reporting to Police
One indicator of public trust in law enforcement is the extent to which people report crimes they experience to police. Data from the National Crime Victimization Survey, conducted annually since 1993, show that reporting rates have remained relatively stable over time (Figure 10). Since 1993, between 40% and 50% of violent crime victimizations and between 30% and 40% of property crime victimizations have been reported to police. In 2024, 48% of violent crimes and 31% of property crimes were reported.
Figure 10. Share of Victimizations Reported to the Police, 1993-2024
Violent Reporting by Race
Reporting rates for violent crime vary somewhat by race and ethnicity, with Black victims generally reporting at higher rates than other groups—often by about 10 percentage points compared to White victims, who tend to have the lowest reporting rates in most years (Figure 11). In 2024, however, reporting rates were similar across groups: 51% of violent crime victimizations involving Black victims were reported to police, compared with 48% for both White and Hispanic victims.
Figure 11. Violent Victimization Reported to the Police by Race, 1993-2024
Clearance Rates
Broader trends in crime, homicide, and arrests provide important context for understanding policing, and they are examined in greater depth in other CCJ publications. For detailed analysis of overall crime patterns, see The Footprint: Tracking the Size of America’s Criminal Justice System; for detailed homicide trends, see Trends in Homicide: What You Need to Know; and for long-term arrest patterns, see Who Gets Arrested in America: Trends Across Four Decades, 1980-2024. Taken together, these analyses show that crime dropped substantially over the past three decades, alongside a corresponding long-term decrease in arrests. Following a sharp increase in homicide during the pandemic years of 2020 and 2021, homicide has fallen significantly, reaching one of the lowest levels in decades.
The crime clearance rate, or the share of reported crimes that result in an arrest or are otherwise cleared, is one indicator of investigative performance. Clearance rates have remained relatively stable over the past three decades (Figure 12). Although homicide clearance rates increased by about nine percentage points between 2022 and 2024, 61% of homicides were cleared in 2024, compared with 65% in 1995.
Figure 12. Violent Crime, Property Crime, and Homicide Clearance Rates, 1995-2024
Officer-Involved Fatalities
No single national system comprehensively documents deaths involving law enforcement, and available data sources vary in scope, methodology, and detail. Information about contextual factors—such as whether the person killed was armed, the number of officers involved, whether charges were filed, and the race, ethnicity, and gender of those involved—is not consistently reported. As a result, each dataset on this topic provides only a partial picture.
Crowdsourced databases generally report higher counts of deaths involving law enforcement. Fatal Encounters4 includes incidents in which law enforcement personnel were present but may not have directly caused the fatality; however, updating of that project ceased in 2021. Mapping Police Violence5 aggregates and verifies information from multiple sources—including Fatal Encounters, the U.S. Police Shootings Database,6 and the now inactive KilledbyPolice.net—and supplements those records with additional research to improve data completeness and accuracy. The Washington Post 7 maintains a separate database that tracks fatal police shootings involving firearms but does not capture deaths caused by other means. While these independent projects show broadly similar trends, their historical coverage is limited.
Government sources such as the CDC’s mortality statistics8 and the FBI’s justifiable homicides statistics9 tend to report substantially lower counts. CDC data rely on finalized death certificates and post-investigation coding, which results in a longer publication lag than incident-based reporting. FBI data capture only homicides classified as “justifiable” by law enforcement agencies and do not include independent third-party validation. As a result, these government series reflect a narrower subset of deaths and should not be interpreted as comprehensive measures of all fatalities involving law enforcement.
Officer-Involved Civilian Fatalities
An examination of all five data sources on police-involved homicides suggests a gradual increase over time, though the pattern varies across sources (Figure 13). Most datasets show a modest upward trend, particularly since the mid-2010s, while FBI data do not reflect this pattern. Estimates of the number of civilians killed in incidents involving police also vary widely across sources, as explained above.
Figure 13. Officer-Involved Civilian Fatalities by Data Source, 1999-2025
By Race and Ethnicity
While the previous chart shows the annual number of deaths in which police were involved, only the Mapping Police Violence data allow for analysis by race and ethnicity. This source indicates that, in a typical year, about 45% of those killed by police are White, about 25% are Black, and about 18% are Hispanic (Figure 14). It also shows that the share of cases with unknown race or ethnicity has substantially increased over time, rising from about 1% in 2015 to 14% in 2025.
When calculated as rates per 100,000 residents within each group, police-involved fatalities differ by race and ethnicity. In 2025, the rate was 6.9 per 100,000 for Black people and 6.1 for American Indian/Alaska Native people, compared to 3.6 for Hispanic people, 2.9 for White people, and 1.1 for those who identify as Asian, Native Hawaiian, or Other Pacific Islanders. These rates use general population denominators and should be interpreted as approximations of relative risk, rather than precise measures of exposure to violence or police contact.
For more information on homicide, see the Council’s Homicide Trends brief. Additional discussion of the limitations of population-based denominators, particularly when exposure to police contact is not accounted for, is available in research on benchmarking racial disparities in police shootings.10
Figure 14. Officer-Involved Civilian Fatalities by Race and Ethnicity, 2013-2025
By Cause of Death
Campaign Zero’s Mapping Police Violence project11 includes information on cause of death, although some cases list multiple contributing causes. Figure 15 presents the primary cause of death. Firearms account for by far the largest share of police-involved fatalities (roughly 87%), followed by motor vehicle-related incidents (about 8%), which include vehicle crashes and pedestrian fatalities involving police vehicles. Deaths resulting from beating or asphyxiation, including neck restraints or other forms of compression used to subdue individuals, represent a very small share of fatalities—about 0.6% of all deaths in 2025.
Figure 15. Officer-Involved Civilian Fatalities by Cause of Death, 2025
Officers Killed or Assaulted on the Job
Officers Killed
Between 25 and 80 officers are fatally assaulted each year, with no clear long-term trend over the past three decades (Figure 16). The number of officers killed on the job peaked at 79 in 1994 and reached a low of 27 in 2013. After increasing 59% between 2020 and 2021, fatalities have remained relatively stable at about 60 per year. Over the same period, the rate of officers killed on the job declined through the 1990s and has largely plateaued at about 0.08 per 1,000 sworn officers.
Figure 16. Officers Fatally Assaulted on the Job, 1987-2024
Officers Assaulted
Between 45,000 and 90,000 officers are assaulted on the job each year (Figure 17). The number of assaults remained relatively stable from the late 1990s through 2019, declined to a low of 47,000 in 2021, and then increased sharply. By 2024, assaults had risen 82% from the low, reaching about 86,000. The rate of officers assaulted declined gradually to a low in 2014 before rising again; in 2025 it reached 111.1 per 1,000 sworn officers, about 74% higher than the 2014 low.
Figure 17. Officers Assaulted on the Job, 1987-2024
Public Opinion of Police
Confidence in the Police
While White and Black victims report crimes to police at similar rates, survey responses about trust in the police differ sharply. In Gallup polls, White respondents consistently report higher levels of confidence in the police than Black respondents, and those differences have persisted over time (Figure 18). Overall confidence peaked in 2004, when 70% of White respondents and 40% of Black respondents said they had “a great deal” or “quite a lot” of confidence in the police. Levels generally declined for both groups in subsequent years, reaching a low of 19% among Black respondents in 2020. By 2025, about half of White respondents and roughly one-quarter of Black respondents reported high confidence in the police.
Figure 18. Confidence in the Police by Race, 1994-2025
The following measures come from Gallup surveys examining Americans’ views of local policing.12 Since 2021, Gallup has tracked public opinion on several aspects of police–community relationships, including satisfaction with local police, beliefs about whether police treat “people like them” fairly, and expectations of respectful treatment. Across these measures, White respondents consistently report more favorable views of police than Black respondents, although the size of these gaps has narrowed modestly in recent years. Attitudes toward police are also more favorable among older adults: Satisfaction and perceptions of fairness and respect increase steadily across age groups, from 18–39 to 40–65 to 65+ (not visualized).
Community Relationships
White respondents consistently report higher levels of satisfaction with the relationship between local police and the communities they serve than Black respondents do, with differences typically around 20 percentage points (Figure 19). In 2024, 81% of White respondents and 64% of Black respondents reported being satisfied with their community’s relationship with local police.
Figure 19. Satisfaction with Community’s Relationship with Local Police, 2021-2024
Fair Treatment
Gallup also asks respondents whether they believe local police treat “people like them” fairly (Figure 20). Among White respondents, the share expressing this belief remained steady at roughly 90% from 2021 to 2024. Among Black respondents, that proportion has been substantially lower and more variable, ranging from 58% in 2022 to 67% in 2024.
Figure 20. Belief that Local Police Treat “People Like Them” Fairly, 2021-2024
Respectful Treatment
Another perspective Gallup has gathered since 2021 is whether respondents believe local police will treat them respectfully (Figure 21). More than 90% of White respondents said they expected respectful treatment, a share that stayed steady throughout the survey period. Among Black respondents, the share was considerably lower in 2021 and 2022, at 69%. By 2024, it remained below the level for White respondents but had increased to 75%.
Figure 21. Belief that Local Police Will Treat Them Respectfully, 2021-2024
Acknowledgements
Policing by the Numbers was originally created in 2021 by Nancy La Vigne, Thaddeus Johnson, and Dan Lawrence as part of the Council on Criminal Justice Task Force on Policing. That work was supported by Ken Griffin, founder and CEO of Citadel; Microsoft; the Joyce Foundation; the Draper Richards Kaplan Foundation; and the National Football League Foundation, as well as the Council’s general operating contributors.
The 2026 update continues and expands upon that original work and was produced by Stephanie Kennedy with support from Carlos Gonzales, Ernesto Lopez, Thaddeus Johnson, Olivia McLarnan, and other members of the CCJ team. This update was made possible through support from CCJ’s general operating contributors.
Suggested Citation
Council on Criminal Justice. (2026). Policing: By the numbers. https://counciloncj.org/policing-by-the-numbers/
Endnotes
1 See Figure 7.1: Kaplan, J. (2025). Decoding FBI crime data. https://ucrbook.com/leoka.html
2 Police Executive Research Forum. (2023). Responding to the staffing crisis: Innovations in recruitment and retention. https://www.policeforum.org/assets/RecruitmentRetention.pdf
3 Leon-Moreta, A., & Fayemiro, O. (2026). Social and political pillars of police diversity: An institutional exploration. Urban Affairs Review, 62(3), 789-829. https://journals.sagepub.com/doi/pdf/10.1177/10780874251364205
4 Fatal Encounters. (2026). https://fatalencounters.org/
5 Campaign Zero. (2026). Mapping police violence. https://mappingpoliceviolence.org/
6 U.S. police shootings data (responses). (2022). https://docs.google.com/spreadsheets/d/1cEGQ3eAFKpFBVq1k2mZIy5mBPxC6nBTJHzuSWtZQSVw/edit?gid=1144428085#gid=1144428085
7 Washington Post. (2024). Fatal force. https://www.washingtonpost.com/graphics/investigations/police-shootings-database/
8 Centers for Disease Control. (2026). National center for health statistics mortality data on CDC WONDER: Underlying cause of death. https://wonder.cdc.gov/deaths-by-underlying-cause.html
9 Federal Bureau of Investigation. (2026). Crime in the U.S. annual reports. https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/downloads
10 Tregle, B., Nix, J., & Alpert, G. P. (2021). Disparity does not mean bias: Making sense of observed racial disparities in fatal officer-involved shootings with multiple benchmarks. In Contemporary Issues in American Policing (pp. 18-31). Routledge. https://digitalcommons.unomaha.edu/cgi/viewcontent.cgi?article=1061&context=criminaljusticefacpub
11 Campaign Zero, 2026.
12 Gallup. (2025). Racial divide on policing narrows 5 years after Floyd death. https://news.gallup.com/poll/690959/racial-divide-policing-narrows-years-floyd-death.aspx
