In 1962, the FBI reported that American police “cleared” 93% of all murders. Today, the national murder clearance rateThe percentage of crimes that police officially close by making an arrest or identifying a suspect. A clearance does not mean a conviction or that the right person was caught. hovers around 60%. At first glance, those numbers tell a simple story: we used to be good at catching killers, and now we are not. But that story falls apart the moment you ask what “cleared” actually meant in 1962. The boss flagged this one for us, and it turns out the answer is more uncomfortable than the question.
What “Cleared” Really Means
A murder is considered “cleared” when police arrest and charge someone, or when they identify a suspect but cannot make an arrest for reasons beyond their control. That second category, called clearance by “exceptional means,” includes cases where the suspect died, fled the country, or where prosecutors declined to press charges. A clearance does not mean a conviction. It does not mean the right person was caught. It means someone was arrested and charged, or that police said they knew who did it.
In the 1950s and early 1960s, this system produced murder clearance rate numbers above 90%. Research from the University of Wisconsin-Milwaukee has noted that DNA evidence now excludes suspects who would have faced a conviction in the 1950s, contributing to the overall decline in clearance. In other words, some fraction of those “solved” cases from decades ago put the wrong person behind bars.
The Era of Coerced Confessions
The legal landscape of 1962 was nothing like today. Miranda rights did not exist until the Supreme Court’s 1966 ruling in Miranda v. Arizona. Before that decision, suspects could be interrogated for hours or days without being told they had a right to remain silent or a right to a lawyer.
The results were predictable. In 1936, the Supreme Court heard Brown v. Mississippi, a case in which three Black men were beaten, whipped, and one was hanged from a tree by a mob that included law enforcement officers. All three confessed. All three were convicted and sentenced to death based solely on those confessions. The Supreme Court unanimously reversed the convictions, but the practices that produced them did not disappear overnight.
As late as 1964, a 20-year-old Black man named George Whitmore Jr. was interrogated for 26 hours in New York and confessed to two brutal murders he did not commit. Manhattan prosecutors later discovered that every detail in Whitmore’s confession was information already known to police. The Supreme Court cited his case by name in the Miranda decision as proof that coercive interrogations produce false confessions.
Research from the Center on Wrongful Convictions at Northwestern found that 84% of documented false confessions occurred after interrogations lasting six or more hours, with the average lasting over 16 hours. The Innocence Project notes that in some cases, people were convicted despite DNA evidence that clearly contradicted their involvement.
The Murder Clearance Rate Decline: Who Paid the Price
The decline was not shared equally. According to the Murder Accountability Project’s analysis of FBI data from 1976 through 2017, the entire national decline in murder clearance rates was driven by cases with Black victims. Clearance rates for homicides with white, Asian American, and Native American victims actually improved slightly over the same period.
A 2024 study published in the Annual Review of Criminology by Philip Cook and Ashley Mancik confirmed this pattern, finding that gun homicides with Black victims accounted for most of the decline. This tracks with what researchers call “case mixThe composition of crime types in a dataset. A shift toward harder-to-solve crimes, like stranger homicides, can lower clearance rates even without any change in investigative quality.” changes: over time, a growing proportion of killings are committed by strangers using firearms, and those cases are inherently harder to solve.
But case mix is not destiny. As The Marshall Project reported, some cities routinely solve two or three times more homicides than others, even after accounting for the types of cases they handle. The variation suggests that resources, community trust, and investigative quality matter enormously.
Fewer Arrests, Better Accuracy?
Here is where the numbers get genuinely counterintuitive. Cook and Mancik’s Annual Review of Criminology study found strong evidence that while clearance-by-arrest rates declined, the likelihood of conviction and actual prison sentences increased. The bar for what counted as a legitimate arrest went up. Fewer people were arrested on thin evidence, and when arrests were made, they were more likely to stick.
Philip Cook, speaking to The Marshall Project, put it plainly: “It also could be that the standards for making an arrest have gone up and some of the tricks they were using in 1965 are no longer available.” Every wrongful conviction story that makes headlines today was once counted as a successful clearance.
A landmark 2014 study in the Proceedings of the National Academy of Sciences estimated that at least 4.1% of all death-sentenced defendants in the United States are innocent. That is a conservative estimate, covering only cases where the threat of execution drove intensive reinvestigation. For ordinary homicide convictions, where defendants receive far less scrutiny, the true rate of wrongful conviction is almost certainly higher.
The Human Cost of “Solving” Cases
The National Registry of Exonerations reported that in 2023, 153 people were exonerated for crimes they did not commit. Nearly 84% were people of color. Nearly 61% were Black. Among homicide exonerations specifically, 85% involved official misconduct: falsified evidence, coerced witnesses, withheld information. The 153 people exonerated in 2023 alone had lost a collective 2,230 years to wrongful imprisonment, an average of 14.6 years each.
These are not abstract statistics. Every one of those exonerations means a real person spent years or decades in prison while the actual killer went free. And every one of those wrongful convictions was, at some point, counted as a “cleared” case.
Where Things Stand Now
The murder clearance rate hit an all-time low of 52.3% in 2022, then recovered to 61.4% in 2024. The improvement coincided with a decline in total homicides, giving overburdened detective units room to work through backlogs.
State lawmakers are beginning to push for better data. Illinois now requires police to publish clearance data on homicides and nonfatal shootings. Michigan, Missouri, and Texas have passed or proposed legislation to boost investigative capacity and improve reporting standards. These reforms acknowledge what criminologists have said for years: the clearance rate, by itself, is a poor measure of justice.
The question is not whether we should solve more murders. We absolutely should. But the nostalgia for a 93% clearance rate misunderstands what that number represented. A system that “solved” almost every case by extracting confessions from people who had no lawyer, no rights, and sometimes no guilt is not a system that was working. It was a system that was efficient at producing the appearance of justice while delivering something far uglier underneath.
In 1962, the FBI reported that American police “cleared” 93% of all homicides. By 2022, that figure had fallen to 52.3%, the lowest on record. The standard interpretation of this trajectory is straightforward: policing deteriorated. But a growing body of criminological research suggests the reality is more nuanced, and more disturbing. The murder clearance rateThe percentage of crimes that police officially close by making an arrest or identifying a suspect. A clearance does not mean a conviction or that the right person was caught. decline may partly reflect improved justice standards, and the high rates of the past may have been built on a foundation of coercion and error. The boss pointed us toward this question, and the evidence turns a simple decline narrative into something far more complicated.
The Mechanics of Clearance
Under the FBI’s Uniform Crime Reporting program, a homicide is “cleared” when at least one suspect is arrested and charged, or when the case is closed by “exceptional means”: the suspect died, fled the jurisdiction, or prosecutors declined charges. Clearance is not synonymous with conviction. It is not synonymous with solving the case correctly. It is an administrative metric that measures police activity, not justice outcomes.
This distinction matters enormously when comparing eras. As John Boulahanis of Southeastern Louisiana University found in his study of Chicago’s clearance data from the 1980s and 1990s, “exceptional clearances” accounted for as much as 20% of the cleared caseload in any given year. Most were labeled “barred to prosecution,” meaning police identified a suspect but prosecutors would not authorize an arrest. These cases were disproportionately likely to involve African-American victims or occur in police districts on Chicago’s South Side.
The Pre-Miranda Legal Landscape
The 93% clearance rate of 1962 was produced by a criminal justice system that would be unrecognizable today. Miranda rights did not exist until the Supreme Court’s 1966 ruling in Miranda v. Arizona. Before that decision, suspects could be held incommunicado and interrogated without being informed of their right to silence or counsel.
The pre-Miranda case law tells its own story. In Brown v. Mississippi (1936), three Black tenant farmers were arrested for the murder of a white planter. One defendant was beaten by a lynch mob that included law enforcement officers, hanged from a tree, and repeatedly whipped. The other two were beaten with a buckled leather belt until their confessions matched what investigators wanted to hear. All three were convicted and sentenced to death based entirely on those confessions. The Supreme Court unanimously reversed the convictions, establishing that confessions extorted by state torture violate due process. But Brown addressed only the most extreme physical violence.
Subtler coercion persisted for decades. In Ashcraft v. Tennessee (1944), the Court found that 36 hours of incommunicado interrogation without sleep was inherently coercive. In Fikes v. Alabama (1957), 10 days of questioning an uneducated man in isolation, denied access to counsel or family, violated due process. In Leyra v. Denno (1954), a psychiatrist specializing in hypnosis was brought in as a “physician” to extract a confession from an exhausted suspect. The Court reversed each of these convictions, but each reversal only set a floor. Everything short of the condemned practice remained available to police.
The case that helped catalyze Miranda illustrates how the system functioned in practice. In 1964, George Whitmore Jr., a 20-year-old Black man, was interrogated for 26 hours in New York and confessed to the murders of Janice Wylie and Emily Hoffert, crimes he did not commit. Prosecutors eventually discovered that every detail in Whitmore’s confession was information already known to police; he had simply been fed the facts during interrogation. The Supreme Court cited Whitmore’s case directly in the Miranda opinion as evidence that “interrogation procedures may even give rise to a false confession.”
The Science of False Confessions
Research has quantified the scope of confessions produced under duress. A study by Steven Drizin and Richard Leo, cited by the Center on Wrongful Convictions at Northwestern, found that 84% of documented false confessions occurred after interrogations lasting six or more hours, with an average duration exceeding 16 hours. The Innocence Project reports that in some cases, defendants were convicted even when DNA evidence contradicted their involvement, because prosecutors constructed theories persuasive enough to convince juries to disregard the biological evidence.
A 2024 integrative review in Behavioral Sciences and the Law identified five distinct typologies of false confessions, noting that most do not involve overt police misconduct. Coerced compliant confessions occur when a suspect capitulates to end the stress of interrogation. Coerced internalized confessions occur when a suspect actually comes to believe they committed the crime, often through suggested blackout scenarios or amnesiac episodes. Both types were common in an era with no recording requirements, no mandatory counsel, and no Miranda protections.
The Great Decline: What the Data Actually Shows
A 2024 study by Philip Cook and Ashley Mancik in the Annual Review of Criminology provides the most comprehensive analysis of what they call “the Great Decline.” Their findings complicate the simple narrative of deteriorating police performance:
The 93% clearance rate of 1962 dropped 29 points by 1994, to 64%. The decline was shared across all regions and city sizes but was not evenly distributed by victim demographics. Gun homicides with Black victims accounted for most of the decline. This aligns with the Murder Accountability Project’s finding that the entire national decline in homicide clearance rates was driven by cases with Black victims. Clearance rates for white, Asian American, and Native American victims actually improved slightly from 1976 through 2017.
Cook and Mancik’s most striking finding concerns the relationship between clearance and conviction. Their preferred explanation for the decline includes an upward trend in the standard for arrest: although clearance-by-arrest rates declined, the likelihood of conviction and actual prison sentences increased. In plainer terms, police made fewer arrests, but the arrests they made were more likely to be correct.
As Cook told The Marshall Project: “It also could be that the standards for making an arrest have gone up and some of the tricks they were using in 1965 are no longer available.” Every person exonerated after a wrongful conviction was once counted as a successful clearance.
The Murder Clearance Rate and the Racial Dimension
The racial dimension of both the old high clearance rates and the modern decline cannot be overstated. The National Registry of Exonerations’ 2023 report found that nearly 84% of the 153 people exonerated that year were people of color, and nearly 61% were Black. Among homicide exonerations specifically, 85% were marred by official misconduct. The 153 exonerees had lost a collective 2,230 years to wrongful imprisonment, averaging 14.6 years per person.
This pattern is not limited to the pre-Miranda era. A 2014 study published in PNAS by Samuel Gross and colleagues estimated that at least 4.1% of all death-sentenced defendants in the United States are innocent, a figure the authors describe as conservative because it captures only cases where the threat of execution drove intensive reinvestigation. For the vast majority of homicide convictions, where defendants receive far less post-conviction scrutiny, the true false conviction rate is almost certainly higher.
Meanwhile, the Murder Accountability Project’s data shows that the modern failure to clear cases falls disproportionately on Black communities. This creates a compounding injustice: historically, Black defendants were disproportionately likely to be wrongfully convicted, and today, Black victims are disproportionately likely to see their killers go unidentified. As researchers have documented, the racial disparity in clearance rates erodes community trust, which reduces witness cooperation, which further reduces clearance rates in a self-reinforcing cycle.
The Structural Explanations
Criminologists point to several structural factors beyond improved legal standards that explain the decline:
- Case mixThe composition of crime types in a dataset. A shift toward harder-to-solve crimes, like stranger homicides, can lower clearance rates even without any change in investigative quality. changes. A growing proportion of homicides are committed by strangers using firearms, as opposed to acquaintances or domestic partners using close-range weapons. Stranger homicides clear at significantly lower rates because they generate fewer immediate leads.
- Declining witness cooperation. In communities with low trust in law enforcement, witnesses are reluctant to provide information. This is both a cause and a consequence of low clearance rates.
- Resource constraints. When homicide rates spike, detective caseloads increase and per-case investigative time declines. The 2020-2022 homicide surge, which saw murders rise nearly 30%, overwhelmed detective units nationwide.
- Forensic limitations. Despite popular assumptions, DNA evidence does not resolve most homicide investigations. Lab backlogs delay results by months, and many crime scenes do not yield usable biological material.
Charles Wellford, emeritus professor of criminology at the University of Maryland, told The Marshall Project that the wide variation in clearance rates between departments shows “it’s not inevitable that there will be low clearance rates.” Investigative quality, resource allocation, and community relationships all make measurable differences.
The Current Picture
The murder clearance rate hit its all-time low of 52.3% in 2022. Since then, it has recovered to 61.4% in 2024, aided by a decline in total homicides that gave overburdened detective units room to clear backlogs. The Stateline reported that the national homicide clearance rate has fallen from 72% in 1980 to 61% in 2024. State legislatures are beginning to respond: Illinois, Michigan, Missouri, and Texas have all passed or proposed legislation to improve clearance rate reporting and boost investigative capacity.
These reforms represent a belated acknowledgment that the clearance rate, by itself, is a poor proxy for justice. A system that clears 93% of cases by coercing confessions from vulnerable suspects is not outperforming a system that clears 60% of cases with convictions that hold up on appeal. The goal is not to maximize the number of people arrested. It is to maximize the number of cases in which the right person is held accountable and the community can trust the result.
The nostalgia for a 93% clearance rate misunderstands what that number represented. It was produced by a system with no Miranda rights, minimal forensic standards, widespread coercion of confessions, and a documented pattern of railroading Black defendants. The decline from that figure is not purely a story of failure. It is partly a story of a system that, slowly and unevenly, began to care whether it was convicting the right people.
That does not mean the current situation is acceptable. Half of all murders going unsolved is a crisis, and the racial disparity in which murders get solved is an injustice compounded over generations. But the solution is better investigation, more resources, and deeper community trust. Not a return to the methods that made 93% possible.
