Wednesday, October 27, 2021

Not that anyone cares, but "violent victimizations" were actually down last year

Your correspondent has not been writing here recently because I'm in the process of evaluating the continued effectiveness of making fact-based arguments on public-policy topics. Texas politics no longer seems to value facts, so I'm not sure what benefit there is from discussing them. Hewing to them just makes one a target and causes one's enemies to believe the opposite out of some twisted, oppositional principle.

And yet, facts persist. Sigh. Let's talk about one that nobody in the political class wants to believe, and that will certainly only bring Grits grief for even mentioning: The just-released National Crime Victimization Survey - one of the two main measurements of crime in the United States - found that, "From 2019 to 2020, the total violent victimization rate declined 22%, from 21.0 to 16.4 victimizations per 1,000 persons age 12 or older." Burglaries and trespassing complaints declined, the survey found. And the decline was especially pronounced among youth age 12-17, for whom violent victimizations decreased 51%.

Does that jibe with the breathless news coverage about crime you've read in the last year? Of course not. The decline in violent victimization won't attract viewers/readers the same way as "Be afraid, killers on the loose," so it won't be reported. Compare this to how the increase in murders played out in the press last year.

As John Pfaff observed on Twitter, this news reinforces the extent to which the reported crime increase being touted near-daily in the press has been largely about homicides. Assaults and violent victimizations overall were down last year, according to this metric, particularly in suburban areas. Simple assaults were down, but also more serious ones. Firearm victimizations, which in this survey includes pointing a firearm as well as firing one, declined from 481,950 in 2019 to 350,460 in 2020. Burglaries were down 19%.

The NCVS is one of two main, national crime measurements in the United States. The "Uniform Crime Reports" (UCR) count crimes reported to law enforcement agencies, while the NCVS surveys crime victims to include crime victims who didn't report to police.

The NCVS is especially important right now because of changes to UCR data. Without going too deep in the weeds, the feds are shifting to an incident-based reporting system that supposedly is more accurate but results in data that's not an apples-to-apples comparison to what was reported before. Plus, many smaller agencies haven't yet made the switch. So for the next few years, it will be problematic to judge crime trends based on these data: They'll be establishing a new baseline that's not comparable to the old numbers.

The NCVS, by contrast, is still performed as it was in the past, though in-person interviews were halted last year for a time bc of COVID. Moreover, the story it tells corroborates UCR data on crime beyond homicides: News reporters and criminologists have puzzled that murders increased but reports of other types of violent crime went down. The NCVS confirms that crime victims simply were victimized less, not just that reporting went down because of distrust of police (or whatever reason one wants to assign).

So many in law enforcement have taken the recent murder spike as an excuse for a money grab, there's little reason to hope public policy might adjust to reflect actual, in-real-life crime data. So Grits offers up the NCVS analysis without any real belief that it will convince anybody of anything. Everyone will just look at it for evidence that confirms their priors, cherrypick that, and move on.

In the context of declining violent crime victimization, it's worth revisiting recent research from Dr. Bill Spelman at the UT LBJ School. He analyzed the major risks of death for Austinites and found murder was a far less salient risk than traffic deaths, suicides, or drug overdoses.

Anyone truly concerned about "public safety" in this town would, by any reasonable measure, be chiefly concerned with the causes of death on the right-hand side of this chart. But we spend most of our time in city politics and a majority of the city budget on the causes on the left hand side. It's exhausting. And stupid. And I don't know if it can or will ever change.

A good start would be for Austinites to vote against Prop A. Early voting is going on right now and election day is next Tuesday. But if you've read this far, you're probably already a Prop A opponent. 


  1. Another read that reinforces Morris Berman's lament of the failure of America. Read this after reading that the Texas legislature has identified 800 books for burning.

  2. Please continue to report. Many support you and appreciate the work that you do.

  3. As I read your report (and I do appreciate your point of view) I could have sworn I was reading an article by a conservative. Why? The left leaning media practices that same tight rein reporting you speak of so eloquently. Hey! We all respect the truth but sadly don’t see it often in print. Good job of doing your homework and for sticking your neck out to have your say. You will allow me to have my own opinion, right? Have a wonderful day. I love Grits for Breakfast!

  4. Scott,

    — Best

  5. The science is continually becoming more clearly established; counterintuitively, facts do Not convince, emotionally driven stories do. What does society do with that ?!

  6. Kind of like all of the emotionally driven "defund the police" arguments in the aftermath of the George Floyd death. Works both ways doesn't it?

    1. I don't know, 17 months later it still sounds like a great idea to me.

  7. No politician cares about facts, only anything that puts the most money in his pocket.

    Keep blogging I read your blog whenever you post anything new.
    Thanks for what you do.

  8. The National Crime Victimization Survey (NCVS) is just that, a survey. Due to the COVID-19 pandemic, the NCVS authors acknowledge they had difficulty obtaining all the 2020 survey data and had to make calculated, weighted adjustments (see pages 3 and 4).

    Instead of relying on a survey, I suggest interested readers go to the applicable FBI website (Crime Data Explorer) and view the actual violent crime data entered by law enforcement agencies across the United States (via UCR and NIBRS - see link below). You will notice there has actually been a significant rise in violent crime over the past several years, including 2020.

  9. Besides the FBI site, I would appreciate it if someone can post a link to updated, reported (not survey) violent crime data. The U.S. economy was locked-down in 2020, so bars, restaurants, clubs, beaches, concert venues, etc. were all, at least partially, closed. In addition, many people started working from home in 2020. Frankly, since there were fewer people out in public in 2020, there were fewer opportunities for criminals to victimize citizens. To that end, I also thought the 2020 violent crime rate would have decreased from 2019 (when the economy was wide open). I'm just trying to find more crime data sources to understand the disconnect...

    Of course, since the U.S. economy is now booming in 2021, more people are back out in public again. Since there are more opportunities for victimization, the violent crime rate this year may end up being significantly higher than 2020. By all accounts, the violent crime rate is up this year:,is%20part%20of%20an%20increase%20in%20domestic%20violence.

    HPD does provide some raw crime data:

  10. Greg, there is no data besides the FBI site and the victim survey. Those are the only two sources. The HPD data you're citing rolls into the FBI data, they're the same thing.

  11. Also the survey reveals LOTS that the FBI data doesn't. Most crimes aren't reported, so you're actually missing a big chunk of what's happening if you only look at reported crimes.

  12. Makes sense. Thanks for the reply. I appreciate it.

  13. Respectfully, I just think suicide is not the same. Part of that is my personal belief, suicide is a reasonable choice in my opinion. But even if one thinks suicide is something society should not allow there are a lot of arguments that it's just not comparable in terms of being problematic re homicide.

  14. I love facts and Grits! Keep it up, Scott. Never give up, pal.

  15. I haven't been here in awhile because I gave up on Texas a couple of years ago and moved to Colorado (of course Denver). The distinction in just about everything from police practice* to living under a government that actually, truly helps its citizens is jarring (plus, they haven't learned about REAL traffic yet; 610 West Loop, 635 LBJ Freeway, or I-35 through Austin on a Friday would show them TRAFFIC). It makes me sad for my homestate and the rational people who haven't left yet.
    That being said, please keep posting. Once we come to believe that the no-nothings, lunatics, neo-authoritarians and other assorted grifter/criminogenic psychopaths are gonna win, the theory becomes self-proving.

    *one of my cars has had only one operational headlight for @ two years; I've never been pulled over. CO has a state income tax, you connect the dots....