I agree with almost everything in this post except for the media criticism parts. The conclusions seem very similar to this January New York Times article, for example:
My sense is that (1) most people believe the spike in murders was related to the Floyd protests, (2) most people believe that because the theory has been widely aired in the media, (3) the people and the media are almost certainly right about this.
What's much harder to say is exactly *how* the protests relate to the murder surge and what could we do about it?
Sorry, the article you link has as its conclusion in that section:
“All three [of the pandemic, more guns, and the BLM protests] played a role,” Richard Rosenfeld, a criminologist at the University of Missouri-St. Louis, told me. “What’s difficult is to assign priority to one compared to the others.”
The thesis of my post here is that this is wrong, and actually the BLM protests were much more important than at least the pandemic (and guns too, although I didn't get into that much because I haven't heard too many people making that case), and that it's quite easy to determine that.
When I said in my San Fransicko review that I thought the BLM protests caused the homicide surge, lots of people said in the comments that they were surprised I believed that because they had heard in the media that it was a collection of factors especially including the pandemic. I think this is the story most ordinary people reading media articles would come away with, and I think it's false. I also linked several articles in this post saying things like that, and why I thought they were wrong.
I agree the media has been willing to say that the protests were one of many potential causes for the surge, but I think they've very much avoided saying they were the main cause, and I think they were the main cause and that this post provides evidence for it.
Scott, I don't think that separating out the contributions of guns is something you've attempted in this post. Obviously the protests were the "cause" of the spike in violence, but it's hard not to think that the abundance of guns (along with the abundance of free time for poor youth) was an inextricable factor feeding into the brutality of the protests, and whether the police are able to de-escalate successfully. I think if we wanted to truly estimate the contribution of guns we would have analyze a lot more data and make a lot more guesses, because I don't know how you could possibly get clear data on this - you would have to find a set of countries with different numbers of guns, but the same racial tensions and a similar trigger event, but I have no idea how you would do that.
Some others here have provided some pretty convincing evidence it might be unrelated. There are already a huge number of guns, so a 3% increase or whatever in the total guns seems like maybe not a big mover. Moreover gun ownership has increased substantially for a generation, yet mostly accompanied a structural decrease in crime. Moreover other local maxima in gun buying did not immediately impact gun crime rates.
I'm not saying that gun ownership was higher during the protests than other times, thus "causing" the violence. I'm saying that the overall high rate might be a precondition for protests leading to riots and looting, and the police not being able to get the situation under control.
It seems to me like all your evidence shows that increases in crime closely followed the death of George Floyd (and Freddie Gray), but not so much that it was because of the BLM protests specifically or that it was because of the factors you mention.
A lot happened in a short time right around Floyd's death, I think it's impossible to separate out. But the obvious explanation to me isn't "BLM protests cause police to do their job less therefore increasing the murder rate, libs owned forever", but more like "high profile examples of police misconduct cause people to trust the police less, call them less often when there is a shooting, feel less of a societal obligation to follow the rules, etc."
In particular I don't think you can so easily separate out these effects:
(1) A high profile police killing, in and of itself
(2) People protesting the killing
(3) Perceived heavy handed police response to the protests
(4) Proliferation of left-wing ideas about police
(5) People not trusting the police so much anymore
(6) Police acting differently as a result of all this
(7) General feeling of a breakdown in order
For your list of proposed mechanisms, recreated below:
> Police interpreted the protests as a demand for less policing, and complied.
> Police felt angry and disrespected after the protests, and decided to police less in order to show everybody how much they needed them.
> Police worried they would be punished so severely for any fatal mistake that they made during policing that they were less willing to take the risk.
> The “Defund The Police” movement actually resulted in police being defunded, either of literal funds or political capital, and that made it harder for them to police.
I think the issue is either these didn't happen as much as advertised, or they've also happened at *other* times.
(1) For "less policing" - progressive DAs pulling back on police activiy has happened recently, but I don't think the increase in crime has correlated with when, or even where, they took office.
(2) For police going on an unofficial work slowdown, this has happened before without the dramatic effect, e.g. under de Blasio in New York
(3) It seemed to me like there was plenty of heavy-handed police action immediately following Floyd, including cops getting off. It also doesn't seem like increases in crime follow *convictions* or decrease following exonerations. At least that's not the obvious pattern in the data you show.
(4) Most places didn't "defund the police" as far as I understand (and if it did I'm guessing it wasn't immediate - budget decisions take time), but the increase in the murder rate seems to have happened all over the place.
This one brought back some menories. In mid-2020, a few of my friends seemed to be convinced that civilization was about to collapse. I live in an extremely safe, wealthy area, so I can't imagine what things would have been like for poor areas in the US.
The general feeling of a breakdown in order really does seem to be precipitated by the pandemic. How many of us have experienced utterly antisocial behavior in the past two years compared with before, in settings that have nothing at all to do with law enforcement?
Thanks ... to be clear this is saying something similar to what I said right? Efforts to reform police behavior don't cause an increase in crime unless they happen after a high-profile police killing?
It is about federal investigations and not popular protests, but directionally similar.
Also, I think the post clearly answers that last question, which was "police withdrawal." Of the four hypotheses that were put forth to explain police withdrawal, if you're looking for policy that could counteract them I'd say the following (respectively):
(1) Fund the police;
(2) Bust police unions and burn away the rot. I think this worked in Newark, if I recall right? Then we can start on the other public sector unions.
(3) Develop clear, inflexible, and bright-line standards for the use of police authority and force that don't depend on general reasonable-person standards, abolish qualified immunity, cap tort damages for a lot of violations and make the punishment firing of the responsible officer(s) rather than some tortuous consent decree or massive monetary liability in the department, mandate that departments carry significant malpractice insurance.
On (3) - I don't think this set of standards exists. Or if it does, it's easy to game. For example, once you lay out an exact set of standards for when foot pursuits are allowed, news will get around and criminals will know when they can just walk away.
You can probably make some improvements on the margin with policy, but ultimately, some kind of general reasonable-person standard is inevitable here. This is unfortunate, for all the reasons you'd expect, but I just don't think there's a better way.
Unquestionably, this is the hardest one. That's why I said "develop" those standards. Of course, you need to retain some flexibility. As it happens, one of the existing standards that I think is usually used appropriately and not normally abused is "exigent circumstances," which would cover a foot chase.
I think the desire for bright lines can be a fool's errand, like in sports. You want well trained professionals with some scope to exercise judgement, and a society and media that doesn't cherry pick bad outcomes as emblematic of the whole.
Absolutely have some data and experience driven guidelines, but the desire to expunge judgement from complex processes generally has pretty poor results. A checklist is great for a plane when the costs of failure are high, situations are consistent, and there is not much time pressure.
Policing isn't so much like that, and particularly the confrontations which are very quick, and quite disparate.
IDK I think at its core people just need to be more comfortable with the idea that police are literally there to be society's enforcers. I think there is a lot of collective/self delusion about that point. Absolutely some things that are currently handled by police are better handled by social work. But sometimes in society someone needs to use force to get compliance, and eggs will be broken occasionally.
The rate of police killings in the US compared to Germany almost exactly tracks how often the suspect is armed in the US versus Germany.
> My sense is that (1) most people believe the spike in murders was related to the Floyd protests
But the thesis here isn't that the spike "was related" to the Floyd protests, a vague assertion that virtually anybody could agree with. It's that the one was principally and proximately caused by the other, that the one followed from the other almost as night follows day.
Suppose we harden the claim to the point of saying that had there been only Covid and gun sales, but no anti-police protests and no breakdown of civil order, there would've been no rise in violent crime at all. Do you still agree that that's perfectly consistent with the general media narrative?
Don't declare certain people essential and others inessential. Don't arbitrarily decide which goods or services may be bought or sold, who is permitted to earn a living and who isn't. Don't declare a universal curfew.
Bad news - our legal system has *always* declared some people essential and others inessential, and arbitrarily decided which goods or services may be bought or sold, and who is permitted to earn a living and who isn’t. The pandemic changed those rules, but didn’t change the basic facts that we have zoning and immigration laws and health and safety regulations.
I think you and Moosetopher agree more than you think you do. If not, I'll stand in and say I agree with both of your points. I think you are right about historical injustices, and I think Moosetopher is right that explicitly saying out loud the quiet parts + changing the rules was a big inflection point that made it acute.
People following Covid restrictions is a disruption -- in my view, and perhaps yours, a tragic one -- of ordinary civil society.
But it's not a breakdown of civil order in the sense I meant that, precisely because people are still following the rules, even if the rules are bad and dumb and possibly even illegitimate at a higher constitutional level.
People smashing windows, setting fires, and throwing bricks at police officers is a breakdown of civil order. And while I suspect that wouldn't have happened without Covid, there are definitely alternate scenarios where Covid happens but that doesn't.
I am not really talking about “covid restrictions.” It was my experience that during the early months of the pandemic, even before restrictions were in place and certainly after, people did not feel free to carry on in their normal habits. Being from the New York area, I know a few people who became sick with Covid in March 2020; even without restrictions, no one I knew was keen on getting infected next. So most of us were glad to radically change our habits to avoid this disease in those first few months. I was living with vulnerable people and went without seeing friends for a whole year until I was vaccinated. I am far from the only one who was living in such an isolated way. But with restrictions or without, it was nevertheless a severe disruption to the civic order that engendered a lot of alienation and distrust that I don’t think has really been fixed. The acute cases of violent protest end and order is relatively quickly restored. But the pandemic in the US has been a disaster for trust in society, and trust is not gained back so quickly.
But I'm not sure it fully explains a punctual spike in violent crime that followed directly upon an episode of violent protest that had police pullback as its explicit, unambiguously desired aim.
Perhaps people taking to the streets to demand de-policing was a sort of derangement that could only have happened under the general civic unraveling that was Covid. But -- to be quite blunt about it -- my civic community was massively disrupted by Covid and *I* didn't get the slightest inclination to smash anybody's storefront or burn anybody's car.
This was a choice that certain people made, and they can and should own its predictable consequences.
> (2) most people believe that because the theory has been widely aired in the media, (3) the people and the media are almost certainly right about this.
Serendipitous to see you commenting here, saves me a link. I'll point at a middle position between this and Scott's, that there is an entire genre of articles in the line of "Black Americans' relationship to the police is complicated, and they're less likely than liberal Whites to support Defund efforts", and that "policing reduces crime" is a pretty standard claim to see defended there. You've written a few yourself, and while I think Scott is wrong to say they don't exist in non-right-wing media it's also not correct to say they're "widely aired" and you might be overestimating the reach.
The causal link between reducing policing and increased crime is a tricky one for plenty of reasons (and unwarranted confidence on the details was my issue with Scott's comment on the previous article), but the basic thesis is at least *available* to anyone who pursues the issue on its own terms
> Nearly a century before George Floyd’s fatal encounter with police at a Minneapolis intersection, a federal agency [redlined] the area a few blocks to the west...
The hilarious things about those Heloc maps people so love to cite, is that they were meant to predict how safe those areas were to lend in, and were in actual fact relatively good predictions (there is obviously some risk here of them being self-fulfilling prophesies, but they were also less influential and more transitory than is commonly presented).
Moreover if you actually examine them, there is just as much or more scourge reserved for "laborers", "servants", "poor farmland", "areas near industry/railyards/quarries", "various "white" ethnic groups", and the overall tone for someone who isn't going into them lookin for "black oppression" is of a more or less earnest attempt to map the expected change in real estate values (and thus suitability for lending) of neighborhoods in what was admittedly a racist era.
The maps were not all made by one person, and some of them definitely seem more or less racially focused, but there is this bizarre narrative that seems to develop that the maps were all about "blacks/negros", and that this then explains the totality of the differential change in the circumstances of different ethnic groups post the 60s. Despite the fact that many many different ethnic groups were "redlined", as any even 3 minute examination of these maps will acquaint you with.
I agree with almost everything in this post except for the media criticism parts. The conclusions seem very similar to this January New York Times article, for example:
https://www.nytimes.com/2022/01/18/briefing/crime-surge-homicides-us.html
My sense is that (1) most people believe the spike in murders was related to the Floyd protests, (2) most people believe that because the theory has been widely aired in the media, (3) the people and the media are almost certainly right about this.
What's much harder to say is exactly *how* the protests relate to the murder surge and what could we do about it?
Sorry, the article you link has as its conclusion in that section:
“All three [of the pandemic, more guns, and the BLM protests] played a role,” Richard Rosenfeld, a criminologist at the University of Missouri-St. Louis, told me. “What’s difficult is to assign priority to one compared to the others.”
The thesis of my post here is that this is wrong, and actually the BLM protests were much more important than at least the pandemic (and guns too, although I didn't get into that much because I haven't heard too many people making that case), and that it's quite easy to determine that.
When I said in my San Fransicko review that I thought the BLM protests caused the homicide surge, lots of people said in the comments that they were surprised I believed that because they had heard in the media that it was a collection of factors especially including the pandemic. I think this is the story most ordinary people reading media articles would come away with, and I think it's false. I also linked several articles in this post saying things like that, and why I thought they were wrong.
I agree the media has been willing to say that the protests were one of many potential causes for the surge, but I think they've very much avoided saying they were the main cause, and I think they were the main cause and that this post provides evidence for it.
Scott, I don't think that separating out the contributions of guns is something you've attempted in this post. Obviously the protests were the "cause" of the spike in violence, but it's hard not to think that the abundance of guns (along with the abundance of free time for poor youth) was an inextricable factor feeding into the brutality of the protests, and whether the police are able to de-escalate successfully. I think if we wanted to truly estimate the contribution of guns we would have analyze a lot more data and make a lot more guesses, because I don't know how you could possibly get clear data on this - you would have to find a set of countries with different numbers of guns, but the same racial tensions and a similar trigger event, but I have no idea how you would do that.
Some others here have provided some pretty convincing evidence it might be unrelated. There are already a huge number of guns, so a 3% increase or whatever in the total guns seems like maybe not a big mover. Moreover gun ownership has increased substantially for a generation, yet mostly accompanied a structural decrease in crime. Moreover other local maxima in gun buying did not immediately impact gun crime rates.
I'm not saying that gun ownership was higher during the protests than other times, thus "causing" the violence. I'm saying that the overall high rate might be a precondition for protests leading to riots and looting, and the police not being able to get the situation under control.
Oh for sure the overall high ownership rate greater increases the violence involved in general policing, and makes dealing with disorder a lot harder.
It seems to me like all your evidence shows that increases in crime closely followed the death of George Floyd (and Freddie Gray), but not so much that it was because of the BLM protests specifically or that it was because of the factors you mention.
A lot happened in a short time right around Floyd's death, I think it's impossible to separate out. But the obvious explanation to me isn't "BLM protests cause police to do their job less therefore increasing the murder rate, libs owned forever", but more like "high profile examples of police misconduct cause people to trust the police less, call them less often when there is a shooting, feel less of a societal obligation to follow the rules, etc."
In particular I don't think you can so easily separate out these effects:
(1) A high profile police killing, in and of itself
(2) People protesting the killing
(3) Perceived heavy handed police response to the protests
(4) Proliferation of left-wing ideas about police
(5) People not trusting the police so much anymore
(6) Police acting differently as a result of all this
(7) General feeling of a breakdown in order
For your list of proposed mechanisms, recreated below:
> Police interpreted the protests as a demand for less policing, and complied.
> Police felt angry and disrespected after the protests, and decided to police less in order to show everybody how much they needed them.
> Police worried they would be punished so severely for any fatal mistake that they made during policing that they were less willing to take the risk.
> The “Defund The Police” movement actually resulted in police being defunded, either of literal funds or political capital, and that made it harder for them to police.
I think the issue is either these didn't happen as much as advertised, or they've also happened at *other* times.
(1) For "less policing" - progressive DAs pulling back on police activiy has happened recently, but I don't think the increase in crime has correlated with when, or even where, they took office.
(2) For police going on an unofficial work slowdown, this has happened before without the dramatic effect, e.g. under de Blasio in New York
(3) It seemed to me like there was plenty of heavy-handed police action immediately following Floyd, including cops getting off. It also doesn't seem like increases in crime follow *convictions* or decrease following exonerations. At least that's not the obvious pattern in the data you show.
(4) Most places didn't "defund the police" as far as I understand (and if it did I'm guessing it wasn't immediate - budget decisions take time), but the increase in the murder rate seems to have happened all over the place.
> (7) General feeling of a breakdown in order
This one brought back some menories. In mid-2020, a few of my friends seemed to be convinced that civilization was about to collapse. I live in an extremely safe, wealthy area, so I can't imagine what things would have been like for poor areas in the US.
The general feeling of a breakdown in order really does seem to be precipitated by the pandemic. How many of us have experienced utterly antisocial behavior in the past two years compared with before, in settings that have nothing at all to do with law enforcement?
“Air rage” was an example that was heavily discussed.
Roland Fryer had already investigated the topic of "viral" police incidents and their effect on crime:
https://marginalrevolution.com/marginalrevolution/2020/06/policing-the-police.html
Thanks ... to be clear this is saying something similar to what I said right? Efforts to reform police behavior don't cause an increase in crime unless they happen after a high-profile police killing?
It is about federal investigations and not popular protests, but directionally similar.
Yeah the “collection of factors” canard is just obfuscation to preserve pious libtard narrative control. It’s obviously 90+% BLM-caused.
Also, I think the post clearly answers that last question, which was "police withdrawal." Of the four hypotheses that were put forth to explain police withdrawal, if you're looking for policy that could counteract them I'd say the following (respectively):
(1) Fund the police;
(2) Bust police unions and burn away the rot. I think this worked in Newark, if I recall right? Then we can start on the other public sector unions.
(3) Develop clear, inflexible, and bright-line standards for the use of police authority and force that don't depend on general reasonable-person standards, abolish qualified immunity, cap tort damages for a lot of violations and make the punishment firing of the responsible officer(s) rather than some tortuous consent decree or massive monetary liability in the department, mandate that departments carry significant malpractice insurance.
(4) Fund the police.
>I think this worked in Newark, if I recall right?
Camden.
On (3) - I don't think this set of standards exists. Or if it does, it's easy to game. For example, once you lay out an exact set of standards for when foot pursuits are allowed, news will get around and criminals will know when they can just walk away.
You can probably make some improvements on the margin with policy, but ultimately, some kind of general reasonable-person standard is inevitable here. This is unfortunate, for all the reasons you'd expect, but I just don't think there's a better way.
Unquestionably, this is the hardest one. That's why I said "develop" those standards. Of course, you need to retain some flexibility. As it happens, one of the existing standards that I think is usually used appropriately and not normally abused is "exigent circumstances," which would cover a foot chase.
I think the desire for bright lines can be a fool's errand, like in sports. You want well trained professionals with some scope to exercise judgement, and a society and media that doesn't cherry pick bad outcomes as emblematic of the whole.
Absolutely have some data and experience driven guidelines, but the desire to expunge judgement from complex processes generally has pretty poor results. A checklist is great for a plane when the costs of failure are high, situations are consistent, and there is not much time pressure.
Policing isn't so much like that, and particularly the confrontations which are very quick, and quite disparate.
IDK I think at its core people just need to be more comfortable with the idea that police are literally there to be society's enforcers. I think there is a lot of collective/self delusion about that point. Absolutely some things that are currently handled by police are better handled by social work. But sometimes in society someone needs to use force to get compliance, and eggs will be broken occasionally.
The rate of police killings in the US compared to Germany almost exactly tracks how often the suspect is armed in the US versus Germany.
> My sense is that (1) most people believe the spike in murders was related to the Floyd protests
But the thesis here isn't that the spike "was related" to the Floyd protests, a vague assertion that virtually anybody could agree with. It's that the one was principally and proximately caused by the other, that the one followed from the other almost as night follows day.
Suppose we harden the claim to the point of saying that had there been only Covid and gun sales, but no anti-police protests and no breakdown of civil order, there would've been no rise in violent crime at all. Do you still agree that that's perfectly consistent with the general media narrative?
How do you have Covid without breakdown of civil order? In this country at least, they have been one and the same.
Don't declare certain people essential and others inessential. Don't arbitrarily decide which goods or services may be bought or sold, who is permitted to earn a living and who isn't. Don't declare a universal curfew.
Bad news - our legal system has *always* declared some people essential and others inessential, and arbitrarily decided which goods or services may be bought or sold, and who is permitted to earn a living and who isn’t. The pandemic changed those rules, but didn’t change the basic facts that we have zoning and immigration laws and health and safety regulations.
I think you and Moosetopher agree more than you think you do. If not, I'll stand in and say I agree with both of your points. I think you are right about historical injustices, and I think Moosetopher is right that explicitly saying out loud the quiet parts + changing the rules was a big inflection point that made it acute.
These were all done in countries that don't have the same problems that we've been having.
People following Covid restrictions is a disruption -- in my view, and perhaps yours, a tragic one -- of ordinary civil society.
But it's not a breakdown of civil order in the sense I meant that, precisely because people are still following the rules, even if the rules are bad and dumb and possibly even illegitimate at a higher constitutional level.
People smashing windows, setting fires, and throwing bricks at police officers is a breakdown of civil order. And while I suspect that wouldn't have happened without Covid, there are definitely alternate scenarios where Covid happens but that doesn't.
I am not really talking about “covid restrictions.” It was my experience that during the early months of the pandemic, even before restrictions were in place and certainly after, people did not feel free to carry on in their normal habits. Being from the New York area, I know a few people who became sick with Covid in March 2020; even without restrictions, no one I knew was keen on getting infected next. So most of us were glad to radically change our habits to avoid this disease in those first few months. I was living with vulnerable people and went without seeing friends for a whole year until I was vaccinated. I am far from the only one who was living in such an isolated way. But with restrictions or without, it was nevertheless a severe disruption to the civic order that engendered a lot of alienation and distrust that I don’t think has really been fixed. The acute cases of violent protest end and order is relatively quickly restored. But the pandemic in the US has been a disaster for trust in society, and trust is not gained back so quickly.
That's eloquently said and very true.
But I'm not sure it fully explains a punctual spike in violent crime that followed directly upon an episode of violent protest that had police pullback as its explicit, unambiguously desired aim.
Perhaps people taking to the streets to demand de-policing was a sort of derangement that could only have happened under the general civic unraveling that was Covid. But -- to be quite blunt about it -- my civic community was massively disrupted by Covid and *I* didn't get the slightest inclination to smash anybody's storefront or burn anybody's car.
This was a choice that certain people made, and they can and should own its predictable consequences.
> (2) most people believe that because the theory has been widely aired in the media, (3) the people and the media are almost certainly right about this.
Serendipitous to see you commenting here, saves me a link. I'll point at a middle position between this and Scott's, that there is an entire genre of articles in the line of "Black Americans' relationship to the police is complicated, and they're less likely than liberal Whites to support Defund efforts", and that "policing reduces crime" is a pretty standard claim to see defended there. You've written a few yourself, and while I think Scott is wrong to say they don't exist in non-right-wing media it's also not correct to say they're "widely aired" and you might be overestimating the reach.
The causal link between reducing policing and increased crime is a tricky one for plenty of reasons (and unwarranted confidence on the details was my issue with Scott's comment on the previous article), but the basic thesis is at least *available* to anyone who pursues the issue on its own terms
Murder spikes the week after George Floyd dies
NYT: It's complicated
Anything happens 60 years after redlining ended
NYT: Redlining strikes again
You can even remove the middle two lines. From the World Economic Forum: https://www.weforum.org/agenda/2020/07/how-redlining-remains-a-source-of-racial-injustice/
> Nearly a century before George Floyd’s fatal encounter with police at a Minneapolis intersection, a federal agency [redlined] the area a few blocks to the west...
Scooby-Doo but every monster turns out to be a redlined neighborhood.
LOL
The hilarious things about those Heloc maps people so love to cite, is that they were meant to predict how safe those areas were to lend in, and were in actual fact relatively good predictions (there is obviously some risk here of them being self-fulfilling prophesies, but they were also less influential and more transitory than is commonly presented).
Moreover if you actually examine them, there is just as much or more scourge reserved for "laborers", "servants", "poor farmland", "areas near industry/railyards/quarries", "various "white" ethnic groups", and the overall tone for someone who isn't going into them lookin for "black oppression" is of a more or less earnest attempt to map the expected change in real estate values (and thus suitability for lending) of neighborhoods in what was admittedly a racist era.
The maps were not all made by one person, and some of them definitely seem more or less racially focused, but there is this bizarre narrative that seems to develop that the maps were all about "blacks/negros", and that this then explains the totality of the differential change in the circumstances of different ethnic groups post the 60s. Despite the fact that many many different ethnic groups were "redlined", as any even 3 minute examination of these maps will acquaint you with.
You should write your own version of this post! Common knowledge for the win!