Less Frequent, More Deadly
Trends in the Lethality of Crime in 17 U.S. Cities
July 2025
In a series of reports, the Council on Criminal Justice has tracked recent shifts in homicide and other crimes. This brief builds on these publications by exploring the lethality (the number of homicides per 1,000 aggravated assaults and robberies) of violent crime in 17 large U.S. cities between 2018 and 2024. It also features a detailed look at longer-term lethality trends in Baltimore, Chicago, and Washington, DC, and complements a companion report on national lethality trends from 1994 to 2020.
Key Takeaways
- In a sample of 17 large American cities, the lethality of violent offenses increased 31% from 2019 to 2020 and was 20% higher in 2024 than in 2018. Thirteen of the 17 cities had higher lethality levels in 2024 than in 2018.
- In Baltimore, the lethality of violent crime rose in 2020 but had dropped by 2023. Firearm lethality, the number of homicides per 1,000 incidents involving a gun, followed a similar trend, falling below 2020 levels by 2023.
- In Chicago, lethality rose sharply from 2019 to 2020, then dropped in 2021 and 2022. Firearm lethality declined beginning in 2021.
- In Washington, DC, lethality remained relatively stable in 2020 compared to 2019 and did not decline later in the study period. Firearm lethality was also flat between 2019 and 2020, although the rate of homicides involving a gun was lower in 2019 than in 2020.
Lethality Explained
In this report, lethality refers to the number of homicides per 1,000 aggravated assaults and robberies. For example, a lethality of 50 means that there were 50 homicides per 1,000 aggravated assaults and robberies. If an aggravated assault or a robbery turns fatal, it is classified as a homicide and no longer counted as a non-fatal offense. An increase in lethality means that homicides made up a larger share of violent incidents, even if the overall homicide rate was stable. If assaults and robberies drop while homicides remain flat, for example, then lethality will rise.
Firearm lethality is the same measure but refers only to incidents involving guns. To determine trends for Chicago, firearm lethality was computed using only assaults due to a major change in gun robberies during the study period.
Many jurisdictions count only the most serious offense, so a robbery that results in death is recorded only as a homicide. Nationally, about 4% of homicides, with known circumstances, are the result of a robbery. For more details, see Expanded Homicide Data, Table 12, in the FBI’s Crime in the United States 2023, publication.Â
Regression result tables are provided in a supplemental document.
Introduction
The United States has long had a high level of homicide. In the early 1900s, its homicide rate was estimated to be about eight times higher than that of England and Wales, and four times higher than that of Australia.1 Today, the U.S. still has a substantially higher homicide rate than other industrialized nations.2 Comparatively, rates of other violent crimes, such as robbery and aggravated assault, are not typically high by international standards. This has led some to argue that the U.S. does not face a general crime problem but rather a problem of lethality.3
Historically, trends in violent crimes in the U.S. have moved in tandem.4 For example, during periods when robbery declined, homicide also declined. But that pattern began to change around 2017, when robbery and homicide trends diverged.5
This report builds on these observations by examining trends from 2018 to 2024 in lethality in 17 large U.S. cities that consistently reported data on homicide, aggravated assault, and robbery. It defines lethality as the number of homicides per aggravated assaults and robberies. While methods for measuring lethality can vary, this approach highlights shifts in the proportion of violent incidents that result in death.6
All city-level data were obtained from online police department portals that provided monthly, incident-level data. Data were collected in January and February 2025. These figures may—and often do—differ from data subsequently published by individual police departments. Differences in how cities define and count crimes may also affect the reported number of offenses and lethality measures across jurisdictions.
Recent Lethality Trends Across 17 Large Cities
Figure 1 shows average lethality and homicide trends for a sample of 17 large U.S. cities, along with lethality trends for each city. On average, lethality rose 12% from 2018 to 2019 and jumped another 31% from 2019 to 2020. Lethality continued to inch upward in 2021 (+4%) and 2022 (+1%), before decreasing 15% from 2022 to 2023 and dropping another 9% in 2024. By 2024, average lethality across the sample was about 20% higher than in 2018.
In 2018, there were an average of 18.5 homicides per 1,000 aggravated assaults and robberies across the 17 cities. By 2024, that number had increased to 22.2. While lethality and homicide rates generally move together, homicide rates dipped in 2021 (-1%) and 2022 (-4%), creating a slight divergence. Overall, the data show that the likelihood of a violent incident resulting in death increased from 2018 to 2024.
Thirteen of the 17 study cities had higher lethality levels in 2024 than in 2018. Arlington, TX, had the largest increase (169%), although it started from the lowest lethality level in the sample, at 4.4 in 2018, and rose to 12 in 2024, making it the third lowest level among the cities. In contrast, Washington, DC had the highest lethality level among the cities throughout the study period, rising 38% from 41.4 in 2018 to 57.2 in 2024.
Figure 1. Lethality Trends for 17 Cities, 2018–2024
A Closer Look at Three Cities
This section examines lethality and gun-specific lethality trends in Baltimore, Chicago, and Washington, DC, from 2012 to 2024. The availability of consistent data made it possible to take a longer-term look at trends in these three jurisdictions. Lethality and gun-specific lethality are analyzed separately. Firearm lethality is of particular interest because a separate study prepared for the Council on Criminal Justice found that incidents involving guns have become deadlier since the early 1990s.7
While the findings are largely descriptive, this analysis also compares lethality in 2020—a year marked by an unprecedented homicide spike and significant changes to violent crime patterns8—to other years in the dataset. The goal was to determine whether 2020 marked a significant shift in lethality and whether any subsequent changes were also significant.
Each city’s results are presented individually, followed by a summary comparing trends across the three cities. For additional context, homicide and gun homicide rates are displayed in the appendix. Regression result tables are provided in a supplemental document.
Baltimore
Figure 2 shows trends in lethality in Baltimore from 2012 to 2024. Lethality remained stable from 2012 to 2014, rose sharply (by 40%) from 2014 to 2015, then dropped 18% in 2016. From 2018 to 2020, lethality increased again before a decline in 2021. Lethality levels in 2020 were significantly higher than in most earlier years, except 2015 and 2019. Though lethality declined in 2021 and 2022, levels were not significantly different from 2020 until 2023. In 2024, lethality was 16% lower than in 2012.
Figure 2. Baltimore Lethality Trends, 2012–2024
Firearm lethality in Baltimore followed a similar trajectory during the study period. Compared to 2020 levels, firearm lethality was significantly lower in most years before 2018, except for 2013 and 2015. Unlike overall lethality, firearm lethality rose slightly in 2021 before beginning to fall in 2022. Firearm lethality levels in 2023 to 2024 were significantly lower than in 2020. In 2024, firearm lethality was 16% lower than in 2012.
Figure 3. Baltimore Firearm Lethality Trends, 2012–2024
Chicago
Figure 4 shows trends in lethality in Chicago from 2012 to 2024. From 2013 to 2016, lethality rose 68%, then dropped through 2019. From 2019 to 2020, lethality increased 57%. Lethality in 2020 was significantly higher than in all earlier years in the study period. Lethality fell slightly in 2021 (-1%), dropped more sharply in 2022 (-11%) and 2023 (-24%), and fell again in 2024 (-3%). Compared to 2020 levels, lethality was significantly lower in 2022, 2023, and 2024. In 2024, lethality was 23% higher than in 2012.
Figure 4. Chicago Lethality Trends, 2012–2024
Firearm lethality in Chicago followed a more complex pattern (Figure 5). This measure excludes gun-related robberies due to a sharp reporting shift: gun-related robberies in the city spiked nearly 500% from 2019 to 2020.9 The increase started abruptly in January 2020, despite a 2% drop in total robberies that year. This likely reflects a change in how weapon use was recorded for robberies. A similar change in gun-related aggravated assaults was not observed; those crimes increased by 36% during the same period.
From 2012 to 2019, firearm lethality fluctuated within a narrow range (±9%) in Chicago. It then rose 15% from 2019 to 2020. Firearm lethality in 2020 was not significantly different from levels in earlier years, except for 2019, when it was at a low of 108.8 per 1,000 gun-related aggravated assaults. After 2020, firearm lethality began to drop, falling in 2021 (-14%), 2022 (-5%), 2023 (-10%), and 2024 (-6%). In 2024, firearm lethality was 30% lower than in 2012.
Figure 5. Chicago Firearm Lethality Trends, 2012–2024
Washington, DC
Figure 6 shows the trends in overall lethality in Washington, DC, from 2012 to 2024. Lethality increased every year from 2012 to 2021, except in 2016. The largest single-year increase was a 57% jump from 2014 to 2015. From 2019 to 2020, lethality increased by 22%, followed by another 14% increase in 2021. It then declined slightly in 2022 (-3%) and 2023 (-8%), before a 5% uptick in 2024. Lethality in 2020 was significantly higher than in all earlier years, except for 2019. However, lethality levels observed in 2021 through 2024 were not significantly different from those in 2020. In 2024, lethality was 341% higher than in 2012.
Figure 6. Washington, DC Lethality Trends, 2012–2024
Figure 7 shows that firearm lethality followed a generally similar pattern, with one exception: While overall lethality increased 14% from 2020 to 2021, firearm lethality decreased slightly (-3%). Firearm lethality in 2020 was significantly higher than in all earlier years, except 2018 and 2019. In the years following 2020, only 2023 had significantly lower firearm lethality levels than those observed in 2020. In 2024, firearm lethality was 183% higher than in 2012.
Figure 7. Washington, DC Firearm Lethality Trends, 2012–2024
Comparative Insights Across Cities
Taken together, the results detail shared patterns and distinct local trajectories across the three cities. Like most cities in the 17-city sample, all three jurisdictions examined over the longer term saw an increase in lethality from 2019 to 2020. Baltimore and Chicago also experienced earlier spikes: Baltimore in 2015 and Chicago in 2016, aligning with a modest national rise in homicide during that period. In contrast, Washington, DC, saw lethality rise almost continuously from 2012 to 2021. In addition, lethality in Washington was higher in 2024 than in 2020, while it was lower in the other two cities.
By 2022, lethality had declined in all three cities compared to 2020 or 2021—a pattern not seen consistently across the broader 17-city sample. Firearm lethality generally mirrored overall lethality in Baltimore and Washington, DC. In Chicago, however, firearm lethality generally declined beginning in 2012, with brief increases in 2016 and 2020.
Conclusion and Implications
The most consistent finding of this report was that the lethality of crime in 17 large American cities increased during 2020. This may not seem surprising given the nation’s historic, one-year increase in homicide from 2019 to 2020. But the rise in lethality shows that homicides increased at a greater rate than aggravated assaults and robberies. Additionally, most of the 17 cities in the sample had higher levels of lethality in 2024 than in 2018, and the average lethality across the sample in 2024 was well above 2018 levels. A closer look at Baltimore, Chicago, and Washington revealed that the 2020 lethality increase in those cities was significant compared to prior years. And, although each of the three cities experienced some drop in lethality over the study period, the declines do not follow the same pattern.
The findings suggest that U.S. cities are not experiencing a generalized violent crime surge, but rather a rise in the lethality of violence. Disputes and crimes that once ended in threats, a scuffle, or injury are increasingly likely to end in death. Unfortunately, this pattern is not new. As one scholar put it nearly a decade ago, “The United States has become more peaceful, but its disputes have become deadlier.”10 While lethality may be experiencing recent declines in some cities, limited evidence suggests that gun carrying may have increased in 2020 and into 2021.11 Further research should investigate more recent trends in gun carrying to better inform policy.
As with all research, these results should be interpreted with care. Aggravated assaults and robberies are subject to underreporting to law enforcement; the sample of cities is relatively small, not representative of the entire nation, and covers a limited period; and it is unclear in the data used for the study how a firearm was involved in gun-related incidents (e.g., brandished vs. discharged).
Gaining understanding of what drives the shifts identified in this report—whether it’s access to guns, gaps in emergency response, changes in policing, changes in community contexts, or a combination of such factors—is critical to designing effective public safety strategies. Continued monitoring and analysis, improved data quality, and targeted research will be essential to guide effective interventions and reduce the deadly toll of violent crime in American cities.
Acknowledgments
Ernesto Lopez was the lead author of this report, with support from Stephanie Kennedy and other members of the Council on Criminal Justice team.
Graduate Research Assistant Bobby Boxerman at the University of Missouri—St. Louis assisted with data collection.
The author would like to thank Ethan Rodgers for reviewing an earlier version of this report.
This paper was produced as part of the work of the CCJ Crime Trends Working Group with support from the Annie E. Casey Foundation, Arnold Ventures, the Harry Frank Guggenheim Foundation, Southern Company Foundation, Stand Together, and CCJ’s general operating contributors.
Suggested Citation
Lopez, E. (2025). Less frequent, more deadly: Trends in the lethality of crime in 17 U.S. cities. Council on Criminal Justice. https://counciloncj.org/less-frequent-more-deadly/
Appendix
Figure A1. Baltimore Homicide and Gun Homicide Rates, 2012–2024
Figure A2. Chicago Homicide and Gun Homicide Rates, 2012–2024
Figure A3. Washington, DC, Homicide and Gun Homicide Rates, 2012–2024
Endnotes
1 Roth, R. (2009). American homicide. Harvard University Press.
2 See Table 1. Kazemian, L. (2022). Long sentences: An international perspective. Council on Criminal Justice. https://counciloncj.foleon.com/tfls/long-sentences-by-the-numbers/an-internationalperspective
3 Zimmering, F & Hawkings, G. (1997). Crime is not the problem: Lethal violence in America. Oxford University Press.
4 Blumstein, A (2000). Disaggregating the violent crime trends. In A. Blumstein & J. Wallman (Eds.), The crime drop in America (pp. 13-41). Cambridge University Press.
5 Roman, J. (2025, January 7). A forecast for less violence in 2025. External Processing. https://johnkroman.substack.com/p/a-forecast-for-less-violence-in-2025
6 The companion study released with this report separates non-fatal violence lethality and robbery lethality. The non-fatal violence lethality includes simple assaults. The authors of that study rely on the National Crime Victimization Survey for non-lethal offenses and can reliably include less serious (simple) assaults without concern about low levels of reporting victimization to police.
7 The gun-involvement data include incidents in which a gun was fired as well as incidents in which a gun may not have been fired. The gun-specific offense indicates that a gun was involved in the incident, such as a robbery in which a gun was brandished but not fired. The data used in this report do not allow for a deeper analysis of how the gun was used in the incidents.
8 This is accomplished by using a logistic regression on violent crime incidents (classified as homicide, aggravated assault, and robbery) for the years 2012 to 2024. The outcome variable is homicide; the analysis coded homicide incidents as one and other violent incidents as zero. The predictor variables were dummy variables for the years 2012 to 2024, with 2020 as the reference year. The analysis repeated this process for firearm lethality.
9 Though it is unlikely to have a meaningful effect, the Chicago incident data do not identify if a gun was involved in a homicide. Instead, gun homicides were generated from homicides as a result of a shooting and were collected from the city’s Violence Reduction database, which can be accessed here: https://data.cityofchicago.org/Public-Safety/Violence-Reduction-Victims-of-Homicides-and-Non-Fa/gumc-mgzr/about_data
10 Berg, M. T. (2019). Trends in the lethality of American violence. Homicide Studies, 23(3), 262-284. https://doi.org/10.1177/1088767919849643
11 National Academies of Sciences, Engineering, and Medicine (2023). Crime rates during the COVID-19 pandemic: Proceedings of a workshop—in brief. Washington, DC: The National Academies Press. https://doi.org/10.17226/26920
