A well-established belief in criminology is that the certainty of being caught is a more powerful deterrent for a criminal than the potential punishment.
So it stands to reason that improving the murder clearance rate — the percentage of murders that are solved — should decrease the murders in an area over time.
Perhaps surprisingly, the numbers don’t back up that idea.
A look at the data shows no relationship between a change in clearance rates from year to year and a change in murder at either a city or national level. And yet the conventional wisdom about the importance of solving murders to reduce murders may still be true.
In part, this disconnect may exist because clearance rates aren’t great statistics to begin with. Also, because murder tends to be clustered geographically within a city, which murders are solved may matter more than the overall percentage.
A failure of measurement?
Thomas Abt, a senior research fellow at Harvard and the author of “Bleeding Out,” a new book on reducing gun violence, points to three reasons that solving murder might be connected to reducing murder:
“First, incapacitation — killers can’t kill again if they’re behind bars. Second, deterrence — if getting caught is more likely, potential killers will think twice. Third, retribution — if killers are brought to justice by the state, then friends and associates of the victim won’t be tempted to take the law into their own hands.”
Abt said in an email: “We don’t have much evidence showing that clearing cases reduces homicides not because it isn’t true, but because it hasn’t been carefully studied. I expect that if we did the research, the relationship between more clearances and less murders would become apparent.”
According to the conventional wisdom, police departments that improved their murder clearance rates from 2015 to 2016 should have been more likely to see a decrease in murders from 2016 to 2017. But for 178 cities that reported substantial data in that period, there was no apparent relationship between percent change in murder clearance rate and percent change in murders.
There is similarly no relationship between solving a high percentage of murders over a longer period and reducing murder. There were 92 law enforcement agencies that reported more than 10 murders and one murder cleared to the FBI each year from 2000 to 2017. At the end of that period, agencies with sustained clearance rates more than 60% for several years (in some cases over a decade) were just as likely to have more murders as they were to have fewer murders.
Take, for example, murders and murder clearance rates in Charlotte, N.C., from 2013 to 2017. Both increased over that span.
Clearance rates are calculated by taking the total number of offenses solved in a given year either through arrest or exception. (Exceptions are when the perpetrator has been identified and located but cannot be arrested — like in a murder/suicide, or when a perpetrator is in jail facing separate murder charges in another state.)
To produce an agency’s clearance rate, that figure is divided by the number of murders that occur in a jurisdiction in a given year.
But the figure for murders solved includes murders that occurred in previous years that are solved in the current year. Additionally, the FBI does not ask law enforcement agencies to break down which cases are cleared by arrest and which are cleared by exception, so there’s no way of knowing how many arrests were actually made by an agency.
This frequently leads to departments that report more murders cleared than occurred in a given year, including Atlanta’s 103% clearance rate in 2017, and an 800% clearance rate reported by the El Paso County Sheriff’s Office, also in 2017.
Reports to the FBI are sometimes incomplete. Murders in New York City and Chicago made up 6% of the U.S. murder total between 2002 and 2017, but the New York Police Department reported clearance data in only six of those years, and the Chicago Police Department does not report any clearance data to the FBI.
There also seem to be errors in the data. In 2012, New Orleans reported to the FBI a murder clearance rate of 15%. But a review of the New Orleans Police Department’s internal homicide tracking spreadsheet obtained through a public records request shows the department probably cleared at least 45% of the city’s 193 murders that year. (We have both worked for the New Orleans department as crime analysts.)
Data on clearances in a city can also only tell us so much because murder and murder clearances can be heavily geographically and demographically concentrated within a given city. The Washington Post reviewed almost 55,000 murders in America’s largest cities over the past decade. Nearly 26,000 of those murders were unsolved, and about 75% of the victims were African Americans in these unsolved cases.
In Boston, for example, The Post’s analysis found that 89.5% of cases with white murder victims had been solved — but only 41.6% of the cases with black murder victims. Which murders are solved by a police department and where may matter just as much, if not more, than the total number of murders solved.
A key appears to be improving the public’s faith in law enforcement, particularly in African American communities. As Abt puts it in his book, “Ineffectiveness leads to illegitimacy in the eyes of the community, so low clearance rates translate into less community trust and confidence in the police.”
In Jill Leovy’s book “Ghettoside,” she describes much of the cause of murder in America as “a problem of human suffering caused by the absence of a state monopoly on violence,” meaning that people who lack service from the police wind up policing themselves.
The random results of shootings
There’s another reason to wonder about the granularity and trustworthiness of the available data. In most cases when a shot is fired, there is no victim. A person is struck by gunfire in only a small fraction of all gunfire episodes, and fewer than a third of all shooting victims die (in most cities for which data is available).
If a police department solves only fatal shootings at a high percentage, a significant segment of gun violence will remain unsolved. Perhaps understandably, more effort is placed on solving fatal rather than nonfatal shootings, but each may be equally important, because whether an individual shooting ends in a death (or not) is largely a matter of luck.
Ultimately, better and more widely available data on murder clearances is needed to shed light on the relationship between solving more murders and preventing them from occurring.