The Sixty-Year Trajectory of Homicide Clearance Rates: Toward a Better Understanding of the Great Decline

Abstract

In 1962, the FBI reported a national homicide clearance rate of 93%. That rate dropped 29 points by 1994. This Great Decline has been studied and accepted as a real phenomenon but remains mysterious, as does the period of relative stability that followed. The decline was shared across regions and all city sizes but differed greatly among categories defined by victim race and weapon type. Gun homicides with Black victims accounted for most of the decline. We review the evidence on several possible explanations for the national decline, including those pertaining to case mix, investigation resources, and citizen cooperation. Our preferred explanation includes an upward trend in the standard for arrest, with strong evidence that although clearance-by-arrest rates declined, the likelihood of conviction and prison sentence actually increased. That result has obvious implications for the history of policing practice and for the validity of the usual clearance rate as a police performance measure. Expected final online publication date for the Annual Review of Criminology, Volume 7 is January 2024. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Explaining the Extraordinary Decline in Chicago’s Homicide Arrest Rates, 1965 to 1994 and Beyond: Trends in Case Mix Versus Standards for Arrest

“…(“Clearance rates” are defined somewhat differently than “arrest rates,” but there is substantial overlap.) A recent analysis of this national trend serves as a companion to this report (Cook & Mancik, 2024). That analysis considers several possible explanations for the phenomenon, dubbing it the “great decline.” It is useful to conduct the same inquiry for Chicago; for one thing, we can utilize unique high-quality microdata, allowing a more reliable and fine-grained inquiry than is possible with available national data.…”

“…The fact that nationwide (and also in Chicago) the prevalence of domestic homicides trended downward starting in the early 1960s, while the prevalence of gang killings increased, suggests that the average arrest rate would have declined even if the police performed equally well over time in investigating these two types of homicide. But the changing case mix, while real, does not appear to account for the national decline (Cook & Mancik, 2024). Attention to case mix in analyzing clearance rates is nonetheless justified, and indeed a logical place to begin.…”

“…For much of this period, there are national data on the likelihood of a prison sentence given arrest; since most convictions (more than 90%) for murder and non-negligent manslaughter receive an active prison sentence, prison-admission data are almost equivalent to conviction data. The prison–arrest ratio trended strongly upward from 1970 to 2000 (Beck & Blumstein, 2012; Cook & Mancik, 2024; Travis et al, 2014), suggesting that the quality of arrests was increasing during this period. That result is a reminder that arrest is an intermediate step in the criminal justice system; to be fully successful, an investigation must end with a conviction.…”

“…Among these characteristics are the victim–offender relationship, and the motivation for the killing. Since such items are not missing at random, they cannot credibly be used as an unbiased predictor of investigation outcome (Cook & Mancik, 2024; Trussler, 2010). For example, indicators for gang involvement or intimate partner violence may reflect rather than just predict the success of the investigation.…”

“…It is logically possible that the trend in the clearance rate is an accurate reflection of the performance of police investigations, and that there really was a sharp decline—although that seems unlikely given the rapidly increasing budgets 14 and major technological innovations during that period (Keel et al 2009; Lum et al, 2022; McEwen & Regoeczi, 2015; Schroeder & White, 2009). The companion article (Cook & Mancik, 2024) explored several other possible explanations for the national trend. It seems plausible that the ultimate explanation for Chicago’s version of the great decline has much in common with the explanation that emerged from analysis of national data.…”