Thank Phil and thanks so much for all of your work on this topic!
I'd be curious to know the degree to which standardization of data reporting played a role in falling clearance rates. The first state UCR programs didn't come about until the early 1970s and they weren't widespread until the 1980s. I'm not sure how one would show that relationship other than temporally, but I've been curious as to whether it played a role.
Very fascinating breakdown in the bias inherently assumed in "clearance" and the alternatives given also with their limitations clearly laid out is fantastic work! Just as I was interested in the topic I found this article. Great job!
This is interesting. On a related note, I have been wondering lately about the extent to which clearance rates have been impacted over time by changes in (1) law enforcement strategy and/or (2) underlying offending behavior. For example: (1) if LE become more strategic about focusing on high frequency offenders, then they would presumably make fewer arrests but each of those arrests would have more crime-fighting benefit. and (2) if offending behavior becomes more concentrated among fewer people (https://www.ppic.org/publication/are-younger-generations-committing-less-crime/), it seems like this could make the number of people arrested per crime committed go down. I think these questions might be more relevant to property crimes (like car theft where anecdotally it seems not uncommon for one individual to commit several crimes before being arrested) than murder. It may come down to how many crimes LE agencies are able to connect the arrestee to. I would hypothesize that there are diminishing marginal returns for LE to connect them to the 10th car theft when they've already connected them to enough to prosecute, for example. But i don't know enough about clearance data and rates. Any thoughts or answers appreciated.
Interesting stuff -- clearly you are on to something here.
I wonder if there is not another effect that should be taken into account. Your model assumed a given percentage of crimes cleared. But in actual departments, the clearance work scales primarily with available personnel. One detective can solve so many crimes a year, no matter how many are committed.
If the model instead holds the "number of crimes solved" constant, you will also see a drop in clearance rate when crime goes up, and a climb again when it goes down.
Thanks Jeff, for the valuable observation about short-term movements in clearance rates for murder -- as defined by the FBI. Over the longer run there are other concerns. Anyone who looks at your chart for murder clearance rates since 1970 will wonder about the remarkable drop from around 90% in 1970 to the 50% or so now. My research with Ashley Mancik identifies the culprit -- that the drop was due to changing standards for making an arrest, and not to declining performance by police investigations.
PJ Cook and AM Mancik. The Sixty-Year Trajectory of Homicide Clearance Rates:
Toward a Better Understanding of the Great Decline. Annual Review of Criminology 2024. 7:59-83
Thank Phil and thanks so much for all of your work on this topic!
I'd be curious to know the degree to which standardization of data reporting played a role in falling clearance rates. The first state UCR programs didn't come about until the early 1970s and they weren't widespread until the 1980s. I'm not sure how one would show that relationship other than temporally, but I've been curious as to whether it played a role.
Good stuff as always, Jeff!
Very fascinating breakdown in the bias inherently assumed in "clearance" and the alternatives given also with their limitations clearly laid out is fantastic work! Just as I was interested in the topic I found this article. Great job!
Or can you email? Mn.safety.alerts@gmail.com
Jeff, can we get your email address? Asking as part of giving Tuesday and would like to support your efforts.
This is interesting. On a related note, I have been wondering lately about the extent to which clearance rates have been impacted over time by changes in (1) law enforcement strategy and/or (2) underlying offending behavior. For example: (1) if LE become more strategic about focusing on high frequency offenders, then they would presumably make fewer arrests but each of those arrests would have more crime-fighting benefit. and (2) if offending behavior becomes more concentrated among fewer people (https://www.ppic.org/publication/are-younger-generations-committing-less-crime/), it seems like this could make the number of people arrested per crime committed go down. I think these questions might be more relevant to property crimes (like car theft where anecdotally it seems not uncommon for one individual to commit several crimes before being arrested) than murder. It may come down to how many crimes LE agencies are able to connect the arrestee to. I would hypothesize that there are diminishing marginal returns for LE to connect them to the 10th car theft when they've already connected them to enough to prosecute, for example. But i don't know enough about clearance data and rates. Any thoughts or answers appreciated.
Interesting stuff -- clearly you are on to something here.
I wonder if there is not another effect that should be taken into account. Your model assumed a given percentage of crimes cleared. But in actual departments, the clearance work scales primarily with available personnel. One detective can solve so many crimes a year, no matter how many are committed.
If the model instead holds the "number of crimes solved" constant, you will also see a drop in clearance rate when crime goes up, and a climb again when it goes down.
Thanks Jeff, for the valuable observation about short-term movements in clearance rates for murder -- as defined by the FBI. Over the longer run there are other concerns. Anyone who looks at your chart for murder clearance rates since 1970 will wonder about the remarkable drop from around 90% in 1970 to the 50% or so now. My research with Ashley Mancik identifies the culprit -- that the drop was due to changing standards for making an arrest, and not to declining performance by police investigations.
PJ Cook and AM Mancik. The Sixty-Year Trajectory of Homicide Clearance Rates:
Toward a Better Understanding of the Great Decline. Annual Review of Criminology 2024. 7:59-83
Valuable insights. Thanks Jeff.