Field Notes

concept

Homicide clearance in the United States

StatusStableReviewdue UpdatedStands on5 sourcesSupportsnotes

A homicide is cleared when police identify a suspect they judge responsible, normally by arrest, or exceptionally when the suspect is dead or beyond reach. The clearance rate — cleared homicides as a share of known homicides — is the closest measurable proxy for the certainty of punishment for murder, and therefore the objective side of Deterrence for the most serious crime. In the United States it has fallen for six decades: from roughly 93 percent in 1962 to a low near 60 percent in the 1990s, a decline of about 29 percentage points by 1994, and it has hovered around 50 to 65 percent since.1 Recent FBI-derived figures put it at 52.3 percent in 2022, 57.8 percent in 2023, and 61.4 percent in 2024.2 Roughly four murders in ten now produce no identified suspect at all.

What drove the decline

Cook and Mancik’s review for the Annual Review of Criminology is the standard synthesis.1 The composition of homicide changed in ways that made it harder to solve: the share of killings committed with firearms rose, gun homicides are cleared at much lower rates than killings by other means, and the circumstances shifted away from domestic and acquaintance killings — which come with an obvious suspect — toward stranger and street killings that do not. The racial composition shift mattered most: homicides with Black victims, disproportionately gun and street killings, grew as a share of the total, and their clearance rate is persistently lower, so composition change alone explains much of the aggregate decline. The review also documents a witness-cooperation mechanism: in high-violence neighborhoods, distrust of police and fear of retaliation suppress the cooperation clearance depends on, which makes the low clearance rate partly self-sustaining.

The racial gap is the story

Since at least 1976, essentially the entire clearance decline has been borne by cases with Black victims. Campedelli’s 2024 exact-matching study, which pairs solved and unsolved homicides on victim demographics, circumstance, and geography, finds cases with Black victims 3.4 to 4.8 percentage points less likely to be cleared than otherwise identical cases.3 MAP’s analysis of FBI data reaches the same conclusion at the aggregate level: the clearance rate for white-victim homicides has been broadly stable, while Black-victim clearance fell and never recovered.2 A homicide detective’s workload, witness environment, and departmental priority all differ by neighborhood, so the gap is not a single-mechanism finding; but its size and persistence mean the certainty of punishment for killing a Black person is substantially lower than for killing a white person, with the deterrent and correction-channel consequences that implies.

Surveillance technology has not moved it

The clearance stagnation coincides with an enormous expansion of investigative technology, and the evaluations of that technology are consistently disappointing. Welsh and Farrington’s Campbell Collaboration review of CCTV found modest crime reductions concentrated in car parks, with no demonstrated effect on violent crime.4 Gunshot-detection systems have been evaluated directly: Doucette and colleagues found no effect on firearm homicide or firearm-related arrests across counties, Chicago’s OIG found ShotSpotter alerts led to evidence of a gun crime in 9.1 percent of deployments and a stop in 2.1 percent, and Piza and colleagues’ 2024 quasi-experimental evaluation found no effect on shootings.567 The pattern matters for how “more policing” is debated: the technologies that scale cheaply — cameras, acoustic sensors, analytics — do not produce clearances, while the inputs that do — detective staffing, time per case, and community trust that produces witnesses — are exactly the expensive ones.

  1. Philip J. Cook and Cameron Mancik, “The Long-Run Decline in US Homicide Clearance Rates,” Annual Review of Criminology 7 (2024): 61-84, doi:10.1146/annurev-criminol-022422-122744. The publisher’s full text blocked automated download; figures cited are from the abstract and the authors’ publicly posted summaries. ↩↩

  2. Murder Accountability Project, analysis of FBI Supplementary Homicide Report and Crime Data Explorer figures, murderdata.org (accessed 2026-08-04). The 52.3 percent (2022), 57.8 percent (2023), and 61.4 percent (2024) rates are MAP’s tabulations of FBI-reported clearances. ↩↩

  3. Gian Maria Campedelli, “Clearing murders: examining the impact of victim race on homicide clearance using exact matching,” Criminology (2024); open-access copy at magazine.fbk.eu. ↩

  4. Brandon C. Welsh and David P. Farrington, “Effects of Closed Circuit Television Surveillance on Crime,” Campbell Systematic Reviews 4, no. 1 (2008), doi:10.4073/csr.2008.17. ↩

  5. Mitchell L. Doucette, Courtney K. Green, Jennifer Necci Dineen, Kerri M. Raissian, and James D. Shapira, “Impact of ShotSpotter Technology on Firearm Homicide and Arrests: A National County-Level Analysis, 1999-2016,” Journal of Urban Health 98 (2021): 609-621, doi:10.1007/s11524-021-00515-4. ↩

  6. City of Chicago Office of Inspector General, “The Chicago Police Department’s Use of ShotSpotter Technology” (August 2021). ↩

  7. Eric L. Piza, Brandon C. Welsh, J. Conor Chalfin, and colleagues, evaluation of gunshot-detection technology deployment, Journal of Experimental Criminology (2024), doi:10.1007/s11292-024-09617-w. ↩

Built on 5 sources (5 external).

Working out connections…