4 ms·

I appreciate your thoughtful comment! We should be careful with what we're talking about here: stops vs. arrests vs. crimes committed vs. offenders. They're al

I appreciate your thoughtful comment! We should be careful with what we're talking about here: stops vs. arrests vs. crimes committed vs. offenders. They're all different things. The argument I'm making is basically this: the murder statistics are the most reliable, so we'll look at those. We should expect a very strong correlation between murder rates and other violent crime, though. And this expectation of much higher rates of violent crime in general should lead us to expect much higher arrest rates among Black Americans relative to other populations all else being equal. Now, all else obviously isn't equal. But the magnitude of the difference in murder rates is absolutely enormous (and it'll about double when we're just thinking about men), so we should expect that the majority of the difference in arrest rates is driven by the same thing...whatever is causing Black Americans to murder at 7x the rate of everybody else. As I commented elsewhere, it isn't actually true that gang-related killings are most killings.[0] If you want to try to paint a picture of the modal murder in the U.S., it's "Young brown man argues with other young brown man about something silly, things escalate, someone pulls out a cheap pistol and shoots the other guy" Gang murders, while profitable fodder for mass media, are a clear minority. 0: https://ucr.fbi.gov/crime-in-the-u.s/2019/crime-in-the-u.s.-2019/tables/expanded-homicide-data-table-13.xls https://ucr.fbi.gov/crime-in-the-u.s/2019/crime-in-the-u.s.-...

cozzyd 6y ago
Perhaps this doesn't generalize outside my city (Chicago), and perhaps data from the CPD is not to be trusted, but here the police claim that ~50% of murders are "gang-related" and 80% are "gang-associated" (i.e. either the suspect or victim is believed to be involved with gangs). The number of suspected gang members is not so high that you would expect 80% of murders to be "gang-associated" by chance (though it's not clear which way the causality goes, it seems highly plausible that people in gangs are more generally violent, but perhaps generally violent people also always join gains). There are also second order effects--more violence around you probably makes you feel more like violence is an acceptable solution?. Note that applying 50% to the number of 2019 Chicago murders already saturates the nationwide "Gangland killings" in the FBI UCR, suggesting that perhaps many of the "unknown" murders would fall into that category. It would also not surprise me if many of the "other arguments" killings are also gang-related or at least gang-associated. Of course, it doesn't help that ~half of murders go unsolved here (and in many other cities) and gang-related violence is especially unlikely to find cooperative witnesses. That said, this is only place with a police department not known for its honesty. Other places might be different. Southern states all have above average white homicide rates, but only Louisiana seems to have an above-average Black homicide rate, at least according to https://www.cnn.com/2018/04/23/health/gun-deaths-in-men-by-state-study/index.html https://www.cnn.com/2018/04/23/health/gun-deaths-in-men-by-s... (caveat 1: this is victims, I could not quickly find corresponding data for offenders. caveat 2: most Blacks live in the South, it's not surprising those states might drive the average.).
- NoImmatureAdHom 6y ago
  Interesting. You've made me revise my estimate of the % of murders that are due to gang violence up. My gut says you're right, Chicago probably doesn't generalize well to the rest of the U.S. It could be that there are huge swaths of the country where there basically are no gangs, and so all murders are not gang-related. But still, the data you're showing makes the federal numbers seem way too low... This [0] suggests that about 1/3 of murders in the U.S. go unsolved. 0: https://projectcoldcase.org/cold-case-homicide-stats/ https://projectcoldcase.org/cold-case-homicide-stats/
  - cozzyd 6y ago
    Well the FBI UCR criteria might be very different from whatever CPD uses. It's amazing how much of an outlier Illinois is in homicide clearance rate...

cozzyd 6y ago

Perhaps this doesn't generalize outside my city (Chicago), and perhaps data from the CPD is not to be trusted, but here the police claim that ~50% of murders are "gang-related" and 80% are "gang-associated" (i.e. either the suspect or victim is believed to be involved with gangs). The number of suspected gang members is not so high that you would expect 80% of murders to be "gang-associated" by chance (though it's not clear which way the causality goes, it seems highly plausible that people in gangs are more generally violent, but perhaps generally violent people also always join gains). There are also second order effects--more violence around you probably makes you feel more like violence is an acceptable solution?. Note that applying 50% to the number of 2019 Chicago murders already saturates the nationwide "Gangland killings" in the FBI UCR, suggesting that perhaps many of the "unknown" murders would fall into that category. It would also not surprise me if many of the "other arguments" killings are also gang-related or at least gang-associated. Of course, it doesn't help that ~half of murders go unsolved here (and in many other cities) and gang-related violence is especially unlikely to find cooperative witnesses. That said, this is only place with a police department not known for its honesty. Other places might be different. Southern states all have above average white homicide rates, but only Louisiana seems to have an above-average Black homicide rate, at least according to https://www.cnn.com/2018/04/23/health/gun-deaths-in-men-by-state-study/index.html https://www.cnn.com/2018/04/23/health/gun-deaths-in-men-by-s... (caveat 1: this is victims, I could not quickly find corresponding data for offenders. caveat 2: most Blacks live in the South, it's not surprising those states might drive the average.).

NoImmatureAdHom 6y ago
Interesting. You've made me revise my estimate of the % of murders that are due to gang violence up. My gut says you're right, Chicago probably doesn't generalize well to the rest of the U.S. It could be that there are huge swaths of the country where there basically are no gangs, and so all murders are not gang-related. But still, the data you're showing makes the federal numbers seem way too low... This [0] suggests that about 1/3 of murders in the U.S. go unsolved. 0: https://projectcoldcase.org/cold-case-homicide-stats/ https://projectcoldcase.org/cold-case-homicide-stats/
- cozzyd 6y ago
  Well the FBI UCR criteria might be very different from whatever CPD uses. It's amazing how much of an outlier Illinois is in homicide clearance rate...