The Incredible Disparity in Firearms Homicide Victimization
The Rise of Pot; I Think the Economy is doing great and so do you!; The False Equivalency, it Burns!; Sleep and Violence; and thoughts on it all
When you boil it all down to its essence, much of public policy comes down to a debate about a simple question: is the scale of the problem to be solved a big number or a small number?
There is an apocryphal story about Mitch Snyder, the Washington DC-based advocate for the homeless who rose to national prominence in the 1980s by picking fights with then-President Reagan in a quest to gain large-scale support for the unhoused.
Among his many successes, Snyder famously was the first to put a number on the unhoused that was broadly internalized across the policy spectrum. Snyder was a character and one who had disdain for “little Western minds that have to quantify everything in sight, whether we can or not.” He likely would not have subscribed to this Substack. But he saw the value in counting the homeless.
The story is that the official count at the time was half a million people unhoused. Snyder thought that the counting process was hopelessly flawed, and that the real number was more like one million. But Snyder knew the power of a big number in a public debate, so he doubled it. Two million.
Now, it is never clear what defines something as a big number or a small number. At some level, big numbers are hard to understand, and small numbers are easy to comprehend. So, if something is beyond comprehension, well, that’s a big number. Two million unhoused Americans at one time is surely a big number.
In that spirit, I would like to share a number with you. A ratio, to be precise.
I was fortunate to work on a paper with Alex Piquero that was recently published in JAMA Open studying racial disparities in firearms victimizations before, during and after the height of the pandemic. Our findings are straightforward. We measured the firearms homicide victimization rate for all individuals, for non-Hispanic/Latino White individuals, for Hispanic/Latino individuals and for non-Hispanic/Latin Black individuals. We then calculated the Black to White (non-Hispanic/Latino) ratio for each of 2018-2022 and for ten age categories.
Source: Piquero AR, Roman JK. Firearm Homicide Demographics Before and After the COVID-19 Pandemic. JAMA Netw Open. 2024;7(5):e2412946.
The core finding is that firearms homicide victimization peaks in most age categories in 2021. And the peak age where the victimization rate is highest is 15-19 years-old.
And the ratio of firearms homicide victimization for 15-19 year-old Black youth is 27 times higher than for White youth.
That’s a big number. It is appalling.
Pot Use in Way Up
Jonathan Caulkins and Keith Humphreys have an interesting piece in Washington Monthly on the huge growth in everyday marijuana use. Their point is that weed is much more potent than in the past, that there is better research on the harms from regular use—particularly for adolescents, and that public policy needs to get moving on this.
These are all fair points. They all contribute new information to the complex cost-benefit analysis that inevitably surrounds any discussion of marijuana policy. In the conventional debate, the costs of use are perceived to be low—especially juxtaposed with the costs of regular use of cigarettes and alcohol which are legal and regulated much as marijuana proponents propose for pot.
But their essay suggests another prior that should be updated in the marijuana cost-benefit analysis. Conventionally, it has always been implied that when access to legal marijuana occurs, some meaningful segment of the new, regular marijuana user population will be substituting pot for alcohol.
So, for the cost-benefit analysis, since marijuana use is mainly safer than alcohol use, if people use marijuana instead of alcohol, that’s a benefit.
Except look at this graphic.
Soaurce: Caulkins, Jonathan and Keith Humphreys. “This isn’t your Father’s Marjuana Use.” Washington Monthly. May, 22, 2024.
What this says is that there are maybe 10 million new, regular users of marijuana since marijuana access started to improve quickly in the mid 2010s. But there are more or less the same number of regular alcohol users. Knowing that is not enough to say that people are not substituting alcohol for marijuana, but it does seem like the graphic above suggests a couple of things are happening:
1. Some people who drank daily are now (also) using marijuana daily, and
2. Some people who did not drink regularly are now daily marijuana users.
The first group are people who drank a lot who now drink a lot and use pot a lot. That’s not great. And the second group includes people who did not drink a lot in the past, who now use marijuana regularly. That’s not great either. I know folks who will argue that for some people, using pot regularly is better than doing nothing, but I defer those folks to Caulkins and Humphreys who discuss that health literature.
So, in summary, there are a bunch of not-so-great bits of news in the article if you are a proponent of marijuana legalization.
I Think the Economy is Doing Much Better, and So Do You!
If you read this blog, you probably don’t spend a ton of time hanging out with macroeconomists. My readers are, I think more of a micro and meso crowd, if I might be so bold as to generalize about you. That is, I think we think more about individuals and their communities and spend less time on GDP. There’s a comparative advantage to specialization, so when it comes to caring about stuff, you kind of have to choose which end of the spectrum to focus on.
But let’s take a minute to look at some macroeconomic data released in the last couple of weeks. Because it shows you are pretty bullish on the American economy. And I am too.
The table above shows survey results from the University of Michigan asking respondents about how they feel about their own finances and the economy as a whole. It’s a little more complicated than it looks—100 isn’t the average level, it’s the 1964 level (which was a pretty great year and most years after have looked sort of bad by comparison).
But anyway. What is clear from this chart is that Americans are substantially more bullish this March then they were last year March. What are they more bullish about? Everything.
So, what does that mean for the fall election? That’s hard to say. The Michigan folks note that from February to March the number of respondents mentioning the election nearly doubled from 13 to 20%, meaning that as we get closer to the election, responses to this survey will be more about partisan affiliation than before, and so less reliable.
But if all things were equal, this should be very encouraging news, at least for everyone who wants a strong recovery.
The False Equivalency, it Burns
Look, I’m not one of those folks excoriating the New York Times on a regular basis for false equivalency in their coverage of all things politics. I do get the Sunday New York Times (we get the Philadelphia Inquirer every day (support your local newspaper people!), the Wall Street Journal weekdays, and The Atlantic, the Economist and the New Yorker—and yes, we hold those physical copies in our hands and look at them through our Costco reading glasses. You should try it, it’s awesome). Plus, I subscribe to a whole bunch of Substacks. So I get that NYT can get a little false equivalency-y here and there. But the writing is wonderful, so mostly I ignore it.
But this article. My Lord. “Even as Violent Crime Drops, Lawlessness Rises as an Election Issue.” The idea of the article is that both Biden and Trump are trying to spin you about whether crime is up or down, and some people believe one side, and some people believe the other side. And you know, fair play all around.
Except the things in the article are just not true.
Here’s the subheader: “In most cities, rates of homicide and violent assault are down significantly from pandemic-era highs. But property crimes have risen, fueling voter anxiety.”
OK, rates of homicide and violent assault ARE down significantly from pandemic-era highs. And that would be sufficient if it was 2023.
But it is 2024. And today, right now, homicide levels are below where they were at the start of the pandemic.
And they aren’t just down a little, they are down unprecedented amounts. So far in 2024, homicides are down more than 40% in Philly, Baltimore, New Orleans, Columbus, Seattle through mid-May. That’s almost half the year. And across all the cities in the data, homicide is down 18.7% in 2024.
Who knows what will happen this summer and beyond, but it is a rock-solid fact that homicide is way down through May, 2024. And I know the reporter can find this information because he quotes the guy (Jeff Asher at AH Datalytics) who produces the data that convinced me that homicide is way down!
How about property crimes? Is the other half of the sub-header introducing the article true? It says, “but property crimes have risen, fueling voter anxiety.” Have property crimes risen?
No!
Here’s the data from the Council on Criminal Justice for 2023, as compared to 2022.
Among property crimes, reports of residential burglaries (-3%), nonresidential burglaries (-7%), and larcenies (-4%) all decreased in 2023 compared to 2022.
Also from the report:
Across the cities [in the report], the residential burglary rate was 26% lower in 2023 compared to 2019.
So residential burglaries are down, nonresidential burglaries are down, and larcenies are down. Where are the property crimes “fueling voter anxiety?”
Ah, here we go, in the fourth paragraph of the NYT article, some specifics.
But property crime, carjackings and smash-and-grab burglaries are up, adding to a sense of lawlessness, amplified on social media and local online message boards.”
Well, I just showed you that nonresidential burglaries are down, which would be the place to record “smash-and-grab burglaries.” So, smash-and-grab burglaries are not up.
What about carjackings? Are they up? Let’s go back to the Council on Criminal Justice, who note that from 2022 to 2023, “[r]eported carjacking incidents fell by 5%.”
None of the specified property crimes are up. Again, I know the reporter had access to this information because he quotes the Council on Criminal Justice.
So, is anything actually up? Yes! Motor vehicle theft has sky-rocketed. And why has it skyrocketed? Is it weak liberal-run governments? Protestors running amok?
No and no.
What is driving the spike in car thefts appears to be car companies re-engineering their vehicles in a way that made them easier to steal. At least that’s the allegation of plaintiffs in several mass torts now consolidated in multi-district litigation in federal courts.
Drivers alleged certain 2011–2022 Kia and Hyundai models don’t have an engine immobilizer—an electronic security device that makes it harder to start a vehicle without the appropriate key. Social media posts that demonstrate how to break in and steal the cars fueled a rash of thefts, the drivers said in consumer deception suits that were combined in multidistrict litigation overseen by Selna.
Ok, so carjackings are down, nonresidential burglaries are down, larceny is down, but somehow, magically property crime is up, right? Because I see in the NYT article that “property crimes have risen, fueling voter anxiety.” Is property crime up?
C’mon, man.
Sleep, and Violence
I came across an interesting study this week, which has an intriguing finding deep in the results section, one that was not the focus of the study. In the study, the Chicago Public Schools (CPS) added a short questionnaire on Adverse Childhood Experiences (ACE) to the Youth Risk Behavior Surveillance System (YRBSS), created by the Centers for Disease Control and Prevention (CDC).
The YBRSS is a national school-based study where,
[T]he YRBS monitors health-related behaviors, experiences, and/or conditions that are significantly influenced by childhood experiences and that contribute to the leading causes of death and disability among youth and young adults.
The ACE measures the “full range of adverse childhood experiences” and the combination of the two in the study measures
Individual and multiple health-compromising, self-soothing behaviors, such as substance use; current experiences, such as perpetrating and experiencing violence; health conditions; and self-reported grades.
The results of the study are that CPS students have a higher prevalence of ACEs than typical school students and the presence of two or ACEs is associated with poor outcomes across a range of dimensions including sexual health and housing insecurity. Black students and LGBQ students were more likely to both have a high number of ACEs and associated poor outcomes.
But there was also a bit of a surprise finding. The study notes,
the majority of all YRBS respondents reported that they get 6 or fewer hours of sleep a night, an insufficient amount of sleep for most adolescents.
This is substantially less sleep than other high school students: CDC reports about 20% of US high school students get less than 6 hours of sleep. There is a ton of research linking a lack of sleep to issues with impulse control and delinquency, particularly in 8-10th graders.
And there is a ton of victimization in Chicago schools. In 2023 there were over 200 serious incidents, including robberies, homicides, sexual assaults, aggravated assaults, and aggravated batteries. Not to mention victimization on school commutes to and from schools and in the community in the late afternoon and evening.
Now, one obvious partial solution to this problem is to have 8-10th graders start their day later. In 2023, California (blue) and Florida (red-ish) became the first states to mandate later start time, starting high school no earlier than 8:30 and middle schools no earlier than 8. There’s a ton of research on the costs and benefits of this policy change, for instance, see this excellent RAND study.
The politics, while bipartisan, are a bit murky. My home school district just went through a drawn-out fight on this question (late school starts won and will begin next year). And the opponents had several good arguments. Parents have morning commitments, including other children, kids have after-school commitments, and transportation is more costly (though it is not obvious to me that this must be true).
And so there are costs, but there are always costs. Nothing in the real-world is Pareto optimal where at least one person is made better off while no one is worse off. And the benefits of better sleep seem overwhelming.
Why not at least give it a test?
And Old Men Dream Dreams
In my corner of the world, these last few weeks have featured three funerals and two surgeries on my own corpus. As you might imagine, my thoughts are expansive these days. The surgeries are an inconvenience; the funerals are an ocean. These were the last relatives of my wife’s parent’s generation. But we have been married for more than two decades. Her people are my people. And we buried her Thea just down from her Papou, my Papou, who I still cannot believe is gone.
But as I process all of this, one vivid image from the long past is indelible.
In 1990, I was a junior at bucolic Kenyon College, dreamily, haphazardly winding my way to some semblance of adulthood. That fall, the college lost an icon, Thomas Greenslade, the school archivist, who died unexpectedly at 80. I worked on the school newspaper and met and interviewed Greenslade several times along the way. He was a lovely person and deeply thoughtful. But a college archivist is a rare choice of icon for an irreverent student body, yet he was. The memorial service was held at the Church of the Holy Spirit in the center of campus, a century-old tower of remarkable massiveness and grace. I went to the service. We all went.
We were a drunken, slobby lot, the Lords and Ladies of the late 1980s, and without cell phones or other modern telecommunications, I cannot recall how we organized. I suppose we read about it in the newspaper. But for the only time in our four years together, we spruced up, sobered up, and showed up, the frat boys freshly together in their khakis and blazers, the young women sparkling in unwrinkled dresses. And we sat together in the nave, really together, as the pews sit perpendicular to the sanctuary and we had only each other to witness.
I’m not sure what I thought would happen there. I assumed, I suppose, that there would be testimonials to Thomas Greenslade’s accomplishments and contributions, that the remembrance of someone who had lived richly would be more celebratory than solemn, that by the time you pass at 80, the time for grieving has gone. But there as we sat quietly and examined each other, and the service moved to conclude, Mary, Thomas Greenslade’s childhood sweetheart and wife of more than fifty years took to the pulpit. “So soon” she wept. “So soon.”
Amen.