Crime

What Explains Why Homicides Are Increasing Significantly Across the Country Since Late May?

A "Minneapolis effect" from lack of policing is a possible explanation for the startling 37% increase in murders in major cities in recent weeks.

|The Volokh Conspiracy |

A new report released yesterday, by Richard Rosenfeld and Ernesto Lopez for the Council on Criminal Justice (CCJ), contains disturbing quantification of what has been reported anecdotally by media: Homicides have increased significantly in many cities across the country since late May.  And the pattern across other crime categories documented in the report suggests that a "Minneapolis effect"—a reduction in policing similar to the "Ferguson effect"—may well be the cause for the recent spike in homicides.

The CCJ report looks at weekly crime data from more than twenty of the nation's largest cities, including New York, Los Angeles, Chicago, Atlanta, and Milwaukee. The report aggregates the data from these major cities and then looks for trends. With respect to homicide rates, the report concludes that:

There appears to be a rough cyclical pattern and a very slight upward trend in the homicide rate over time. The model estimated a structural break near the end of May 2020, after which the homicide rate increased by 37% through the end of June. The rise in homicide was led by three cities: Chicago, Philadelphia, and Milwaukee.

Here is the weekly homicide data depicted as a scatter plot.

The vertical red line indicates a "structural break" identified in weekly homicide rates between January 2017 and June 2020. (The structural break model assumed the break point was unknown and allowed the model to estimate the significant break in the series and adjusted for seasonal effects by comparing to crime rates during the same week in the previous year.)

The report observes the same pattern of a recent increase for aggravated assaults.  Looking at 17 large cities for which data were available, a structural break near the end of May 2020 was detected as well. Aggravated assaults rose by 35% from late May through the end of June 2020. The rise in aggravated assaults was led by Chicago, Louisville, Nashville, and Detroit.

Here is the weekly aggravated assault data depicted as a scatter plot:

Here again, the vertical red line indicates a structural break in the data set—a structural break that obviously exceeds ordinary seasonal variation.

What is noteworthy about the post-late-May spikes in homicides and aggravated assaults is that this same pattern does not appear for all other kinds of crimes. The report looked at nine other crime categories—gun assault (a subset of aggravated assault), domestic violence, robbery, burglary (and also the subsets of commercial and residential burglary), larceny, motor vehicle theft, and drug offenses. None of the other crime categories exhibited a structural break starting in late May. Burglaries, for example, abruptly increased (by 190%) at the end of May and then equally abruptly returned to normal levels in the next week. A dramatic increase in commercial burglaries in the week that coincided with the mass protests following the George Floyd killing explains this pattern.

The crime categories that came the closest to exhibiting the same pattern as homicides and aggravated assaults were gun assaults and robberies.  Gun assaults also showed a sharp and sustained increase after late May, although the pattern was not clear enough to be identified as a structural break. And robbery exhibited a structural break, but the timing was slightly earlier.  Robbery exhibited a long term downward trend, but after March 2020 the robbery rate rose by 27% through the end of June.

I have previously blogged about a recent and similar pattern in crime increases in Minneapolis. In that city, shooting crimes (including homicides) increased after the police killing of George Floyd in late May—but other crimes did not. A key part of the explanation appears to be that Minneapolis police stopped making as many street stops as they made previously. And given the unique responsiveness of gun crimes (such as homicides) to police activity, the result of the reduction in Minneapolis appears to have been a sharp increase in gun violence. At least 275 people have been the victims of gun crimes in Minneapolis during the first seven months of this year, surpassing the entire annual totals of all but two of the past ten years.

A parallel pattern of firearm crime increases occurred during the 2016 Chicago homicide spike.  A detailed paper on the Chicago spike by my University of Utah colleague Richard Fowles and me explains that in Chicago in 2016 gun-related crimes increased dramatically, but not other crimes. (See pp. 1600-01 of the study). Specifically, in Chicago in 2016 homicides increased substantially, by 58% year-over-year from 2015 to 2016.  There were also large (more than 20%) increases in robbery and aggravated assault–but not such large increases in other index crimes. Focusing specifically on gun crimes, there was a substantial increase in shootings in Chicago in 2016. Fatal shootings increased by 66% and non-fatal shootings increased by 44%.  (See Table 5).

Professor Fowles and I explained in our paper that the most likely cause of the 2016 Chicago homicide spike was a reduction in street stops (often referred to as "stop and frisks") in Chicago. We called this effect the "ACLU effect" because an agreement between the Chicago Police Department and the ACLU in in August 2015 was implemented in December 2015, leading in 2016 to about an 80% reduction in street stops conducted by Chicago police officers.

The new CCJ report discusses what might be causing the recent spike in homicides and aggravated assaults. The report notes that while some of the violence was directly connected to protest activities, in most cases the crimes appear to have involved perpetrators other than the protesters. Most of the increase in violent crime took place away from the demonstrations and was not limited to single week (such as the week surrounding May 25, when George Floyd was killed).

The CCJ report also finds it instructive to compare the recent rise in urban violence with the increases that occurred five years ago in the wake of protests about a police killing in Ferguson, Missouri. As the report explains,

Analysts tied the heightened violence to two versions of the so-called Ferguson Effect. The first connects the violence to "de-policing," a pullback in law enforcement. The second essentially turns this explanation on its head and connects the violence to "de-legitimizing," positing that communities, disadvantaged communities of color in particular, drew even further away from the police due to breached trust and lost confidence. As a result of diminished police legitimacy, fewer people reported crimes to the police or cooperated in investigations, and more engaged in street justice to settle disputes. It remains unclear whether either of these theories explains the previous rise in violence, much less today's increase.

The CCJ report is  properly cautious in concluding that we don't know for certain whether de-policing or de-legitimizing policing was responsible for the increase in homicides and other violent crimes in 2015-16 … or which of the two is (apparently) responsible for the recent homicide increases since late May. But my prediction is that a de-policing "Minneapolis Effect" (akin to the earlier de-policing "Ferguson Effect" or the "ACLU Effect" in Chicago) will ultimately possess the most explanatory power for the recent and abrupt homicide spikes.

Here's what the recent data appear to show: In the wake of George Floyd's killing, police have been redeployed to respond to anti-police protests, diverting them from anti-gun patrols and other activities that deter the carrying of illegal firearms. Police have also also pulled back from some measures of pro-active policing, again as the result of the protests. The result of this reduction in law enforcement activity directed against gun violence has been, perhaps unsurprisingly, an increase in gun violence.  Chicago Police Superintendent David Brown put the point precisely for his city, explaining that "[e]very time we have to drain our resources for protests, the people on the West Side and the South Side suffer."

The de-policing hypothesis appears to be a better fit with the data than the de-legitimizing hypothesis. If law enforcement has become generally de-legitimized, the result should be a general increase in all crime categories, including (for example) burglaries. But that is not the pattern that the data exhibit. The recent data reflect an increase in homicides and other gun-related crimes but not other crimes—a unique pattern that is more consistent with de-policing being the cause.

One way of attempting to sort through the competing hypotheses is to examine whether residents of the affected citiies are reporting fewer crimes to the police. In analyzing the 2016 Chicago homicide spike, for example, Professor Fowles and I used calls to 9-1-1 as a measure of police legitimacy, as was suggested by other researchers. We found that changes in 9-1-1 call rates could not explain the Chicago homicide spike. Examining recent 9-1-1 call patterns might be one way to investigate any explanatory power of the de-legitimizing hypothesis.

Similarly, police departments often keep data on street stops and other measures of police activity.  Changes in these activities should be assessed to see if they correspond to increases in homicides. For example, in my recent blog post, I pointed to a decrease in Minneapolis in police searches and suspicious person stops that coincided precisely with the increase in shootings that occurred there in late May. Data in other cities could be reviewed to see if they exhibit similar trends.

If the hypothesis about a Minneapolis Effect is correct, a recommended policy response would seem to be to bolster police activities targeting gun violence. Some modest good news on this front comes from Chicago. Last week, in a widely publicized move, federal law enforcement officers from the Bureau of Alcohol, Tobacco and Firearms and other federal agencies deployed to Chicago, focusing on enforcing federal firearms laws. Federal law enforcement efforts against gun crimes can take advantage of tough federal sentencing laws and avoid some of the pretrial release issues that have drawn attention in Chicago.

And perhaps even more important given its vastly larger numbers, the Chicago Police Department created a new "Community Safety Team" that deployed officers to the South and West Sides of Chicago where the spikes in violent crime are occurring. The Community Safety Team is designed to not only supplement existing law enforcement efforts, but also to engage in regular community projects. While it is too early to tell what the effects will ultimately be, the very early results are encouraging. This past weekend in Chicago, the Chicago Police Department recorded 39 shootings. That compared to 41 the previous weekend, 50 the weekend before that, and 87 during the long Fourth of July weekend.

Researchers should continue to investigate why homicides have been spiking in Chicago and other major cities across the country. If the answer is that de-policing is linked to rising gun violence (as some earlier studies would suggest), further limiting police efforts to aggressively deter gun crimes will tragically lead to more shootings and more homicides. And the victims of those crimes will likely come disproportionately from African-American communities—communities that, in some instances, may want more aggressive police efforts to combat gun crimes.

Attorney Barr touched on these de-policing issues in testimony before the House Judiciary Committee yesterday:

When a community turns on and pillories its own police, officers naturally become more risk averse and crime rates soar. Unfortunately, we are seeing that now in many of our major cities. This is a critical problem that exists apart from disagreements on other issues. … And it is not just that crime snuffs out lives. Crime snuffs out opportunity. Children cannot thrive in playgrounds and schools dominated by gangs and drug pushers. 

The crime rate patterns discussed above support the Attorney General's position. These patterns deserve serious scrutiny as the data continue to emerge. The stakes are obviously very high in determining how the nation should respond to this recent spike in gun violence.

Note: I am a member of the CCJ, but was not involved in drafting of the report discussed in this post.

Advertisement

NEXT: Academic Career Success and the Importance of Ideological Conformity: An Anecdote

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. I am a qualitative rather than quantitative researcher so please bear with me, but take the Chicago funeral mass shooting (15 victims). The Chicago PD had two marked cruisers there — that’s not a diminished police presence.

    Reportedly individuals in a stolen vehicle started firing at mourners who then returned fire, with the police recovering some 60 shell casings from the area afterwards, I’m not sure if that includes casings from the rounds fired within the vehicle.

    What I see here is more of the “they can’t get us all” mentality that destroyed the 55 MPH national speed limit in the 1980’s. I don’t believe that they’ve made any arrests yet, and self defense or not, I highly doubt that any of those weapons were legally possessed.

    The other thing to factor in are the psychological consequences of the quite fascist “lockdowns” that much of the country has been subjected to for the past 4+ months. There is an underlying layer of suppressed anger that doesn’t help mediate disputes.

    Bottom line — there were two marked cruisers and a “tactical unit” which people may or may not have known was present, but marked cruisers are highly visible. To have a drive-by shooting in that environment, along with people shooting back, is more police impotence than a diminished police presence.

    1. ” To have a drive-by shooting in that environment, along with people shooting back, is more police impotence than a diminished police presence.”

      Alternatively, it’s about people who just don’t care if there’s police around or not.

      1. You’re right. The real factor in crime is Democrats not the presence of police.

      2. “Alternatively, it’s about people who just don’t care if there’s police around or not.”

        That’s what I mean by “police impotence” — not only that they don’t care but need not care because there will be no consequences.

        https://www.youtube.com/watch?v=LQIRD57lls8

        1. You missed the point, too.
          Many criminal acts are committed by irrational people. Irrational people don’t think about consequences, because that’s a rational thought process. If you aren’t thinking about consequences, cops or no cops has no relevance.

          1. Is there some reason you suppose there are suddenly a lot more irrational people around?

            Because the alternative is the same number of somewhat less irrational people making a calculation that they’ll get away with it.

            1. Exactly. Ever notice that a whole lot of irrational people otherwise become quite rational when confronted with authority and the possibility of getting the worse end of it. Crime is the result of a weakness or vulnerability — actual or perceived — of the victim by the perpetrator. When elected officials communicate that people won’t be arrested even if they engaged in criminal acts, they are communicating a general weakness or vulnerability. Not surprisingly, criminals — running the gamut from rational or irrational — take their cute from that.

            2. Brett Bellmore wrote: “Is there some reason you suppose there are suddenly a lot more irrational people around?

              Yes there is — or are. (1) The pandemic itself. (2) The social and economic fallout from the pandemic, including job and income losses, everyday activity losses, and so forth. (3) The enormous press play-up of the Floyd murder — and I’m /not/ saying the murder was inconsequential or that it somehow “shouldn’t” have been played up, just that it was a factor.

              All these lead to a circumstance in which individuals or groups are more likely to act not on the basis of their long-term self-interest but on the basis of current emotionality. This can lead to social movements as well — including violent ones.

              I’m /not/ arguing that a great many people aren’t acting in a way that is “rational,”or calculated, based on the new reality. Just that it isn’t the only thing going on.

            3. “Is there some reason you suppose there are suddenly a lot more irrational people around?”

              Is there some reason you’re putting your theory in my mouth?

    2. Police presence is only one factor of diminished policing. If criminals know they are unlikely to be charged even if they are arrested that diminishes the effect of the police force, and, as you say, makes them basically impotent (which in turn leads to James Pollack’s observation, that criminals just stop caring if there are police around or not)

      1. You missed the point.

        Fearing police response and prosecution for criminal acts is a rational process. Irrational people lack rational thought processes. Many criminal acts are conducted by irrational people.

        1. Be very careful of expanding the definition of mentally incompetent into irrational behavior.

          For example, an irrational person may be injured rushing down a staircase while an insane person may think he can fly and will jump out a window instead.

          I don’t think these shooters are insane.

          1. “Be very careful of expanding the definition of mentally incompetent into irrational behavior.”

            Mentally incompetent thinking is irrational. That’s not my doing.

        2. Less than you would think. Or at least not completely irrational.

        3. Fearing police response and prosecution for criminal acts is a rational process.

          What about those that know about and acknowledge the response, but consider this to be low risk? You only need to worry about the consequences if you get caught. What if you rationally think you won’t get caught?

          1. “What about those that know about and acknowledge the response, but consider this to be low risk? ”

            That’s definitionally not irrational if you can explain the rationalization.

        4. If irrational people were the cause then there would be no spike at all since they would have been irrational before and irrational now. Something has to be different beyond chalking it all up to crazy.

          1. “If irrational people were the cause then there would be no spike at all”

            This is an assumption on your part. I don’t pretend to know what makes irrational people act irrationally.

  2. “What Explains Why Homicides Are Increasing Significantly Across the Country Since Late May?”

    Cabin Fever.

    1. That’s my instinct too. You can’t lock people up with no jobs, no socializing, and a boob tube for entertainment, and expect no problems. People aren’t laptops you can close the lid on.

      1. Exactly.

        And once the state loses legitimacy for its fiats on issue A (i.e. lockdowns), it concurrently loses its legitimacy for all its rules.

        1. Most Americans support lockdowns, so far as I have observed.

          Some loud, disaffected, poorly educated citizens have objected to the lockdowns (and to masks, and probably to stop signs), however.

          1. Have you observed most Americans? What a massively inane statement! You haven’t observed many, if you’ve been observing the lockdowns.

            Now, maybe the polls you read, in the media you trust, might have convinced you of such, but the sheer explosion in cases after the lockdowns started getting lifted puts lie to your observations; people, lots of people, were dying to get out! Oh, that or it was the protests, take your pick.

          2. .01 out 10. But on par for your trolling & abuse. No whining about being censored, 1 supposes you hold out for EV, who ignores you.

            1. To be fair, complaining about censorship to anyone but the censor is generally ineffective.
              You should take a cue, and start ignoring the complaints you can’t do anything about.

  3. People in these cities have been promised better education, more jobs, and better social services by Democrats for decades, and seen those promises forgotten the day after the election. They should be shooting people, but for some reason they’re shooting the wrong ones.

    1. No, no, no — merely voting against them will be good enough…

      1. The problem is in most of these cities there is no one else to vote for. Sure there might be other names on the ballot, but from a policy standpoint they might as well all be the same person

        1. True. In D.C. they don’t even have Republican primaries.

          1. A few election cycles back, the R’s didn’t even put candidates in every statewide race. They weren’t winning any of them, which led to a shortage of good candidates, which contributed to the habit of losing all the elections, which further tightened the supply of R candidates willing to lose elections. They offered Bill Sizemore as a candidate for governor, when his name was extremely toxic because he made his living qualifying whackadoodle ballot initiatives for right-wing sponsors. Towards the end of that business, all the liberals had to do to campaign against one of his ballot measures was tie his name to it. Then they tried a basketball-player-turned financial-manager.
            anyways, right before I started law school, one of the profs took a leave of absence to run for Attorney General as a Democrat, He won the contested primary, and surprised everyone by winning the Republican primary, as well, despite not being on the ballot in that one. At the time, Oregon didn’t allow candidates to be listed with more than one party in the general election, so he was listed in the general as a Democrat. He served one term but didn’t go back to the law school afterwards.

            1. Left out that I was talking about Oregon, which in my youth had R’s in both Senate seats and the governorship, but hadn’t managed to occupy any of them in the years before I left.

            2. The 1990 election explains Massachusetts — the Democrat (John Silbur of BU fame) was far more conservative than Republican Bill Weld. Since then, excepting the 8 years of Deval Patrick, MA has had a purported Republican Governor — but the MassGOP largely abandoned all other races.

              When Charlie Baker ran for re-election in 2018, a totally unknown and rather controversial street preacher got almost a third of the votes in the GOP primary.

    2. ” They should be shooting people, ”

      Who should they be shooting?

      (Big-talk, impotent, stomped-by-their-betters right-wingers are among my favorite faux libertarians and favorite culture war casualties.)

      1. We know this already. It’s an ancient trick of politicians to focus your local, disaffected population’s rage on an external enemy, so they don’t focus you.

  4. The most salient thing in that graph isn’t the pandemic spike, it’s that we really like to kill each other every June and July.

    1. The sun stays out later, so people stay out of their houses longer, interacting with each other. Some of them do so poorly, and the guns come out.

      1. The sun stays out later, so people stay out of their houses longer, interacting with each other. Some of them do so poorly, and the guns come out.

        Not even an hour earlier, you blamed the spike on “cabin fever” — i.e., people staying out of their houses less, interacting with each other less. You seem to be grasping at anything at all other than the blazingly obvious “fuck the police” mentality that broke out right at that same time.

        Proof positive that “It is difficult to get a man to understand something, when his salary depends on his not understanding it.”

        1. Cabins get damn hot this time of year…..

          1. “Cabins get damn hot this time of year…”

            There’s this thing called “Central Air Conditioning”. All the cabins in these parts have it.

        2. “Not even an hour earlier, you blamed the spike on “cabin fever””

          For those who can read, I blamed two different spikes on two different factors.

          “Proof positive that “It is difficult to get a man to understand something, when his salary depends on his not understanding it.”
          Clarify for me, upon exactly WHAT does my salary depend?

          1. Or were you referring to YOUR salary here?

  5. It is fairly well established that violent crime spikes in the summer in at least northern cities, particularly murders. This won’t let me post web links but Gifford center amongst others discuss this. So, this may be a signal/noise problem, how much of the spike is attributable to less policing and how much to regular summer uptick? Better comparison may be to last year during same months.

    1. Please look at the graphs in the OP. Specifically the X axis.

    2. The structural break model used in the underlying paper controlled for seasonality by looking to the same week in the previous year. Here is what the study’s authors say about the structural break model: “Because street crimes tend to rise and fall with seasonal changes, the estimates are adjusted for seasonal effects by controlling for the crime rate during the same week in the previous year.” So seasonality does not seem to be the explanation.

      1. Then why does your graph have a linear trend line? That’s a dataviz crime

        1. To show the “very slight upward trend in the homicide rate over time” mentioned in the first line of the extract shown above.

          1. That trend line appears to be a simple linear regression, which would be totally inappropriate for cyclical data. You can’t measure a change over time this way. Did you notice that the data series starts in January, at the low point of the annual cycle, and ends in summer towards the high point? Even if the peak-to-peak trend was downward, cut-off pint shenanigans like that would show a secular increase (which is why it’s a no-no to use linear regression on cyclical data).

    3. I agree with you, and the use of linear recursion on the charts makes looking at some of these a PITA. The plots still seem to show much higher than year over year numbers.

      What shocked the hell out of me was how cyclical some of these are. The aggravated assault rate is a very nice sine wave.

      1. There’s an old joke about the correlation of ice cream sales and the murder rate, implying that ice cream causes more murders

    4. The graph used in the article is dumb for showing how large the shift is. Also his 37% nationwide number is somewhat misleading.

      What is actually happening is that in some cities they are experiencing huge spikes in homicide, seasonally adjusted. NYC and Minneapolics homicides increased over 100% for the month of June compared to 2019. Chicago had increases in the 80%ish range. Milwaukee’s homicides increased 130%, Baltimore had a 10 year high in 2019, which was exceeded in both June and July of this year.

      1. 150 years ago, people who were unhappy living in the city could go west, leaving the people happy in the cities to stay in the cities, happy. Today, people who are unhappy living in the cities have no meaningful alternative destination, because the land in the west already has owners, who react forcefully to trespassers.

  6. I skimmed the report and saw no mention of attempting to normalize the data based on recent spikes in unemployment. Wouldn’t it be impossible to know how much of the spike to attribute to reduction in policing without accounting for other causes in the relevant timeframe?

    1. The structural break in homicides occurs in late May – unemployment began spiking upward many weeks earlier – so unemployment does not appear to be a good candidate for explaining what is happening here. The timing is off.

      1. Can you point me to studies tracking the effects of unemployment on criminality that show “many weeks earlier” is too distant to explain the phenomenon? It seems to me that a person who loses their job probably doesn’t walk out the door and jack a car. They spend some time being frustrated, trying to find ways around not having a job, etc. As that frustration grows (rent becomes due, etc.) they may perform crime.

        I’m willing to accept your theory (“many weeks earlier” is too far) but are you aware of any empirical evidence to support your theory?

        1. Unemployment might be the cause even if timing is off. June is when a lot of people who became unemployed in March are likely to be out of cash.

      2. “The structural break in homicides occurs in late May – unemployment began spiking upward many weeks earlier – so unemployment does not appear to be a good candidate for explaining what is happening here. ”

        Replace “running out of cash reserves” for “unemployment”, then. Some people have significant savings to draw upon if the lose their income, while others are closer to desperation.

        1. I’d guess very few who are inclined towards murder desist because they still have cash on hand. Jean Valjean just stole bread…

          1. Not all murders are for food and shelter. A lot of domestic homicides result from tension. I’m told that sometimes a spouse losing a job and not being able to contribute to the house can increase marital tension. As Gale Snoats said in Raising Arizona:

            “And as per usual, I wouldn’t be surprised if the source of the marital friction was financial.”

          2. “I’d guess very few who are inclined towards murder desist because they still have cash on hand.”

            Perhaps, although desperation is known to alter peoples’ thought processes.

            “Jean Valjean just stole bread…”

            Because that’s what the author chose. the danger of choosing a fictional example is that they’re, well, fictional…
            So when I go to write the story it’s a bit different from Mr. Hugo’s version, ol’ Jean goes down to the corner market intending to just steal a little bit of bread, but gets caught by the shopkeeper. Jean, feeling put upon and embarrassed about being reduced by circumstances to need to steal to feed the family, resists being held by the shopkeeper, and a struggle ensues. One thing leads to another and pow! there’s the shopkeeper lying on the floor, a pool of blood slowly expanding around the dent in his head. Jean valJean flees into the night, scared and hungry…

  7. Liberals have wiped out 20+ years of progress in reviving the American City all in the matter of a few months. Anyone who thinks that liberalism is not extremely dangerous should think back to like what New York City looked like in 1986 and then compare that to what it will look like in 2021.

    1. What do you think NYC will look like in 2021?

      1. Trump-free. Says he’s a Florida resident now…

      2. Probably the bombed out shell it was in 1986. Once it becomes clear businesses are not moving back, the tax base is going to disappear overnight, and urban blight will take over. Look at pictures from the late 60’s compared to the early 70’s of any American city. Stark despite the fact it was only a few years. That is only going to be amplified by the fact everything moves faster now.

        1. America has been moving in the other direction — the backwaters emptying, the smart and ambitious people choosing urban and suburban communities, the rural areas declining to the point of dysfunction — for decades. This seems destined to continue, with bright flight intensifying.

          1. People WERE migrating back to cities. We will see if Covid changes that pattern. I think it is to early to tell, but the increase in crime and far left liberals running cities isn’t going to help the recovery.

            1. ” We will see if Covid changes that pattern.”

              In the long term? Unlikely. Infectious disease used to be much more common in cities, then we learned why that was and engineered around it. Eventually, the medical professionals developed treatments and immunization. We’re at peak antibiotic. As the disease pathogens continue to develop resistance to the antibiotic arsenal we have, THAT might make a difference.

      3. Well, my wife and I are certainly thinking of moving out, if crime continues to increase. So my prediction is for a city that is poorer and more crime-ridden. And that votes even more Democratic. Paradise for some.

        1. The long-term trend for crime in America is down. If we wind down the War on Drugs, that trend will improve that much more.

  8. I think it may be related to the anger over the killing of George Floyd. I suspect there are several factors including general anger leading some people to be less able to control their impulses, a sense that things are worse for some people, dispair and frustration.

    1. I don’t want to sound pessimistic, nor do I wish to tread on individual grief, but there’s a “George Floyd” pretty much every year.

      1. More than one based on my reading of the Short Circuit summaries.

      2. “I don’t want to sound pessimistic, nor do I wish to tread on individual grief, but there’s a “George Floyd” pretty much every year.”

        Which is a big part of what is driving the anger over it.

  9. I see that after people pointed out that Prof. Cassell’s stop-and-frisk theory was fatally flawed because it relied on completely ignoring NYC and handwaving why that was justified, he… just went back to repeating the same theory.

    1. David – I don’t believe your point is a fair one. In the original Chicago paper, Professor Fowles and I devote five pages to why New York City was exceptional and could not serve as a model for Chicago (and presumably other cities). See pp. 1608-13. So we did not “ignore” New York, as you suggest.

      1. “Ignoring” and “came up with a reason to discount it” seem like distinctions without a difference.

        I’m curious why this CCJ organization decided it would be a good time to do this study, and if it would have been published if the trend was slightly downward rather than upward. Perhaps someone familiar with that organization can point to me a CCJ paper that runs counter to the political leanings of its members.

        1. Seems like a man who doesn’t see a difference between explanations for something and ignoring it isn’t very curious about anything beyond his own navel.

          It can’t be that a timely look at negative trends occurring to get at why they’re occurring might be informative to reverse them (same goes for positive, but that’s less urgent), no gotta be some nefarious bias at play.

          I’m at a point where you should be given what you demand and then some. Defund the police entirely, wall off the cities and force the denizens deal with the consequences of their choices. How long did CHAZ stay a “summer of love”?

          1. “Defund the police entirely, wall off the cities and force the denizens deal with the consequences of their choices.”

            It sure sound like you’re a “very curious” person.

            There is no dispute that Cassell ignored the New York example. He explained his reasoning, but that doesn’t mean he included it in his data. In any event, DMN said “and handwaving why that was justified”. Everybody knows that Cassell explained his exclusion of NYC.

          2. “It can’t be that a timely look at negative trends occurring to get at why they’re occurring might be informative to reverse them (same goes for positive, but that’s less urgent), no gotta be some nefarious bias at play.”

            People with strong ideological identification tend to see everything in terms of their ideology. This is, of course, not a problem to other people who share that ideology.

            1. Neither is it a problem for folks with the opposite strong ideology, gives them the excuse to write off the entire thesis without needing to engage with the facts. Bonus points, it gives them the joy of moral outrage.

              To the detriment of all.

              1. “Neither is it a problem for folks with the opposite strong ideology”

                gosh, I hope that doesn’t give you justification to blow the whole thing off without engaging with any facts. Meanwhile non-partisans get to laugh at both. While still being stuck with having to choose between nominees able to get their party’s backing.

  10. Two things here.

    Stop and frisk is a proactive disarming of demographics that are overrepresented in shootings. Less frisking, more guns, more shooting. Simplistically, if the cops had been stopping and frisking at the funeral, there would have been no return fire.

    Isn’t there a known correlation between the economy and crime? Doesn’t a prolonged downturn, especially one concentrated among higher crime demographics, lead to more violence?

    1. What’s wrong with return fire?

      1. The problem with return fire is the stray rounds going downrange…

        1. The problem with return fire is noncombatants stopping stray rounds from going downrange.

    2. Collective guilt? Getting rid of individualized suspicion and going with what your demographic profile says?

      Not only is that a pretty immoral policy, as was noted above there’s evidence it didn’t even work.

      1. There’s no systematic racism without collective guilt. Glad to hear you’re rejecting that.

        1. Systematic racism is not about guilt.

          Stop and Frisk was not systematic racism; it operated as intentional racism.

          1. So if “systematic racism is not about guilt” will you reject as unjust any remedial actions that are predicated on anyone being guilty of anything related to systematic racism?

            Just wondering. I look forward to quoting you.

            1. Yeah – affirmative action predicated on the collective guilt of white people is not a thing I support.

              Recognizing and mitigation systemic distortions in our meritocracy does not require any beneficiaries to be guilty either collectively or individually.

              1. Please point me to an explanation, designed for high school students, of how why some of them will get into better colleges than others solely because they have same skin color as long-dead people who were once slaves. Maybe you should write it!

                1. How we judge merit in academia – grades, standardized tests, legacy, etc. doesn’t really test the merits of folks who have resource differentials.
                  Part of that resource differential is racially based. Not just the legacy of slavery, but that of Jim Crow and redlining and the criminal justice system and all sorts of other stuff. It’s not just racial, but it does have a particularly racial component.

                  Recognizing that there is talent that academia is not seeing and working to draw it out is a good idea short-term since it’ll get us a more talented workforce, and long-term because it’ll make the next generation have a more level playing field and thus that our country has a better meritocracy, to ensure a more talented workforce at that point.

                  1. Leaving aside the veritable forest of begged questions in there, I should like to focus on your upliftingly aspirational second paragraph.

                    Let us look at three individuals :

                    A, who has had past resources of 100, is capable of achieving, with duly sprinkled 50 units of academic magic dust, merit level 100, but with no magic dust only 70.

                    B…resources 120 ….magic dust 70…. merit level 100 -150

                    C…resources 55…. magic dust 120 ….merit level 70 -110

                    We have 120 units of magic dust to spare. Should we sprinkle in on

                    (a) A and B; or

                    (b) on C ?

                    (a) will yield merit of A=100,B=150, C=70 ; total = 320
                    (b) will yield merit of A=70, B=100, C=110; total = 290

                    We therefore have a conflict – raising up the disadvantaged in prior resources produces a less talented (as measured in merit) workforce. Though it does narrow the differences in merit.

                    There’s no guarantee (at all) that equalising past disadvantage will help the village overall. Nor for that matter, is there any guarantee that even if we somehow managed to provide everyone with equal resources, the village would finish up with the best workforce possible. It’s much more likely that concentrating resources on the most talented children would maximise talent in the workforce.

                    If you can spend $25 to get child A from merit level 60 to 80, and another $25 to get child B from merit level 100 to 150, great. But what if spending the whole $50 on child B would get it from 100 to 200 ?

                    There’s no reason to believe that equalising educational opportunity will maximise the village’s potential. Still less concentrating resources on the poorer performers.

                    So we need to be clear about the goal. Do we want to maximise the aggregate potential of the village, or do we want to minimise the difference between the highest and lowest performing villagers ?

                    1. “Nor for that matter, is there any guarantee that even if we somehow managed to provide everyone with equal resources, the village would finish up with the best workforce possible. It’s much more likely that concentrating resources on the most talented children would maximise talent in the workforce. ”

                      Donald Trump got into the Wharton business school. You much magic academic dust did that take?
                      His acceptance kept an honest person out.

                      All the whining about affirmative action in college admissions doesn’t stand up to that.

                2. “Please point me to an explanation, designed for high school students, of how why some of them will get into better colleges than others solely because they have same skin color as long-dead people who were once slaves.”

                  File it next to the paper where the people descended from the slave-owners can get into better colleges because their families have inherited wealth.

      2. “Collective guilt? Getting rid of individualized suspicion and going with what your demographic profile says?”

        Isn’t that the basis for all affirmative action and all collectivism as a whole? Caring more about a person’s statistical profile and what groups they fit in than about the individual’s details?

        1. Affirmative Action has many potential rationales. Collective guilt has been pretty roundly rejected as a policy or legal matter.

          For me, the point of affirmative action is that, as currently gathered, an individuals’ details present a distorted view of an individuals’ merits. We’re leaving talented people behind because our playing field is tilted.

          1. “We’re leaving talented people behind because our playing field is tilted.”

            You’d expect the people who can’t compete unless the rules operate in their favor to complain that the rules are otherwise fair.

            I can’t jump as high as professional basketball hoops. The NBA is biased against me. The rims should be at 5′ 6″!!! And who let all these foreigners in here? Tell you what, pay me a league minimum salary and I’ll go sit down here at the end of the bench and coach won’t hear a peep of complaint out of me the whole season. Although if you let me out of quarantine I might get caught “buying food” down at the nudie bar.

        2. “Isn’t that the basis for all affirmative action and all collectivism as a whole?”

          It is if you’re against it.

  11. If police will not protect the citizens of a city, organized crime will step in and do the job. The will offer protection to the citizens at a price.

    1. I thought you conservatives favored private enterprise over public employees. If someone can provide protection at a better price than the police, shouldn’t the people be free to choose which one offers them protection?

  12. I think I’d want to have alternative explanations investigated as well, before marrying any particular theory.

    One example: early on I read somewhere that the cartels get some of their drug precursors from China, and the manufacturing/shipping disruptions were causing a drug shortage, so street prices were rising. That could certainly lead to more robberies by addicts, as well as gangs fighting over turf or ripping each other off, which usually leads to murders.

    To be clear, this might be hogwash, it was just Some Dude on the Internet, but it’s not quite enough to find one plausible thing that correlates with whatever you are interested in; you also need to discredit other plausible causes by e.g. looking at street drug prices and availability.

    1. “To be clear, this might be hogwash, it was just Some Dude on the Internet”

      Indeed.

      1. Well, here you go: “The surge in drug-related violence partly comes down to changes underway in global markets. The slowdown in international trade means that cartels and gangs have fewer opportunities to move their illegal merchandise through global supply chains, whether by land, air, or sea. Meanwhile, drug producers in Bolivia, Colombia, Peru, and Mexico are struggling to source precursor chemicals to make methamphetamines, fentanyl, and even cocaine. The lockdown of Chinese factories is not only impacting the export and sale of iPhones but also of counterfeit goods.”

        Source. Lots of similar stuff available if you google “pandemic cartel drug prices” or similar terms.

        1. Same dude on the Internet.

  13. I don’t claim to know, but when life is good for people I think they’re enjoying living it instead of focusing on whatever might motivate them toward violence.

    Things were very good until March. Then the virus hit. Governments instituted inhumane restrictions in an extreme overreaction — because they didn’t know any better at the time. Over the next few months, people learned enough to make their own decisions, but governments are still in inhumane overreaction mode.

    So life is distinctly not good for people. Government inhumane overreaction continues, with no end in sight. When you treat people inhumanely, you shouldn’t expect them to remain peaceful.

    And even if government stopped making it worse, life would not be close to as good as it was before March. It should be no surprise that people focus their negative emotions on each other.

    1. ” Governments instituted inhumane restrictions in an extreme overreaction — because they didn’t know any better at the time.”

      “Stay home if you can. Wash your hands”… inhumane.

      1. Putting people out of work and keeping them out of work is inhumane. Taking away childrens’ education is inhumane.

        If they only made requests, then no problem. Mandates are not requests.

        Congrats on showing the people suffering that you don’t care about the hardships imposed on their lives, BTW.

        1. You find being required to help keeping other people healthy to be inhumane and you’re lecturing me about not caring about people. Shove off, buttwipe.

  14. I appear to have stumbled into a meeting of Libertarians For Abusive, Bigoted Policing.

  15. I’m not really seeing a “structural break” in the homicide plot as there are only 7 data points above the trendline. Looking back to about May-June of 2017 there is a series of about seven data points above the trendline before one point below followed by about seven more above. There also appears to be a “structural” dearth of data points below the line about May-June of 2019. Perhaps if you put some imaginary threshold at a rate of 0.5 then yes there are more data points above but then the slope of the trendline is also positive so it looks more like a slight statistical anomaly rather than a “structural break” to me.

  16. Large scale increases in unemployment likely are having an effect on violent crime. Too many young men not gainfully employed.

    And not enough entertainment venues are open. End the damn lockdowns and a lot of this goes away.

    1. “End the damn lockdowns and a lot of this goes away.”

      How does ending the damn lockdown suddenly give people money to spend on/at entertainment venues?

      Or are you just making a demand, and we’re supposed to assume to meant to end with “or I’ll go start shooting people”?

  17. Did anyone look at numbers for arson? Because the riots are a better explanation than anything related to Covid-19 (though the relaxation of laws against masks may very well have helped the people doing the looting, pillaging, and burning to get away with it).

    1. The report did not examine arson; the burglary increases in late May seem to represent the looting during riots.

      1. Are lootings reported as burglaries?

        1. Different jurisdictions may report crimes slightly differently in these data and there is no official (from the Uniform Crime Report) definition of looting so burglary is the closest. And give the abrupt, large spike of burglaries, even when examining the data on a daily level, the spike in burglaries coincides with media coverage of protests/riots/looting on the same nights.

          1. “burglary is the closest.”

            Except it isn’t. Depending on the actual circumstances, robbery or shoplifting are the actual crimes. Burglary involves entering peoples’ dwellings. Most states have statutorily amended this requirement to include places of business. At common-law, burglary involves invading someone’s home under cover of night, and that invasion of home is part of what makes the crime a felony.

            1. If you read the report there was a large spike in commercial burglaries, which coincided with the demonstrations. Larceny, which includes shoplifting, went down. Robbery did have a small increase in more recent months; residential burglary went down. And we are not talking about common law. Every state has a criminal code for burglary or, as I mentioned in my comment, uses a standardized definition of burglary which is provided by the FBI to code their data.

Please to post comments