Episode 14: When the Police Back Off

There have been many questions about police proactive behavior and its impact on crime.  In 2020 two events caused police officers to “back off” on their street activity.  Dr. Jessica Huff explains the research into de-policing and how a reduction in proactive behavior lead to an increase in some types of crime.

Main Topics

• Police officers backed off on their proactive behavior as the result of both department policy during COVID, as well as the social pressure after the death of George Floyd.
• Street-level officers demonstrated a significant reduction in some behavior.
• Backing off on proactive behavior was associated with an increase in some types of violent and property crime.

If you are interested in the de-policing research discussed today, you can find the article using the following citation:

Nix, J., Huff, J., Wolfe, S. E., Pyrooz, D. C., & Mourtgos, S. M. (2024). When police pull back: Neighborhood‐level effects of de‐policing on violent and property crime, a research note. Criminology, 62(1), 156-171.

Don't forget to like, FOLLOW, and share. Sharing this podcast or an episode is one of the best complements I can receive, which will help grow the show.

And don't forget to provide a review. Giving five stars is never a bad idea.

Feel free to email me your comments using the "send us a text" option (above), or at the following email address: policeinservicetrainingpodcast@gmail.com

You can also contact me at: Bluesky: @policeinservice.bsky.social

The views and opinions expressed in this podcast are those of the author and guests, and are not authorized by and do not necessarily reflect those of the New York State Division of Criminal Justice Services or the State of New York.

Transcript

[00:00.000 --> 00:16.240]  Welcome to the Police in Service Training Podcast.
[00:16.240 --> 00:20.720]  This podcast is dedicated to providing research evidence to street-level police officers and
[00:20.720 --> 00:22.340]  command staff alike.
[00:22.340 --> 00:26.640]  The program is intended to help the police and law enforcement community create better
[00:26.640 --> 00:31.880]  programs, understand challenging policies, and dispel the myths of police officer behavior.
[00:31.880 --> 00:34.980]  I'm your host, Scott Phillips.
[00:34.980 --> 00:40.160]  In the early to mid-1990s, there was a significant increase in the crime rate across the United
[00:40.160 --> 00:41.280]  States.
[00:41.280 --> 00:46.800]  For example, in 1990, there were over 2,000 murders in New York City alone.
[00:46.800 --> 00:51.420]  An explanation for this was the drug market, specifically cocaine and crack, which led
[00:51.420 --> 00:56.560]  to competition where drug dealers killed each other for a larger share of the market.
[00:56.560 --> 01:01.920]  By the end of the 1990s, crime in the U.S. started to fall at a fairly steady rate, and
[01:01.920 --> 01:06.360]  by 2020, it had fallen roughly 50% across the United States.
[01:06.360 --> 01:11.640]  In 2023, New York City had only 390 homicides.
[01:11.640 --> 01:16.760]  As most people likely know, crime spiked again about four, maybe five years ago.
[01:16.760 --> 01:21.140]  Recent research offers two explanations for this spike.
[01:21.140 --> 01:24.800]  The first is COVID, and the second is de-policing.
[01:24.800 --> 01:30.600]  That is, the police backed off on proactive behavior after nationwide protests had resulted
[01:30.600 --> 01:32.400]  from the death of George Floyd.
[01:32.400 --> 01:38.040]  Clearly, these two things are different, but they are possible to untangle.
[01:38.040 --> 01:41.740]  Any disruption in the natural order of things can lead to questions.
[01:41.740 --> 01:47.160]  In this case, both COVID and the George Floyd incident were disruptions that, as a matter
[01:47.160 --> 01:50.740]  of fact, contributed to the police backing off.
[01:50.740 --> 01:54.640]  So the natural question that comes next is, what happened to crime?
[01:54.640 --> 01:55.640]  Did it go up?
[01:55.640 --> 01:56.640]  Did it go down?
[01:56.640 --> 01:58.320]  Or did nothing happen?
[01:58.320 --> 02:02.260]  Answers to these questions can help the public and the police understand the potential impact
[02:02.260 --> 02:05.860]  of planned reductions in police activity.
[02:05.860 --> 02:11.240]  To better understand the idea of de-policing, we are joined by Dr. Jessica Huff.
[02:11.240 --> 02:16.480]  Jessica is an assistant professor in the School of Criminal Justice at the University of Cincinnati.
[02:16.480 --> 02:21.920]  She uses experimental methods to examine programs and policies intended to improve police effectiveness
[02:21.920 --> 02:23.420]  and fairness.
[02:23.420 --> 02:27.900]  She has also partnered with numerous police agencies to evaluate different programs, including
[02:27.900 --> 02:33.260]  body-worn cameras, intelligence-led policing to gun crimes, and the use of force.
[02:33.260 --> 02:34.260]  Welcome to the podcast.
[02:34.260 --> 02:36.060]  Thank you for having me.
[02:36.060 --> 02:37.060]  Okay.
[02:37.060 --> 02:41.380]  So when we met at a conference in Albany, I think it was back in September, I was asking
[02:41.380 --> 02:46.500]  you about your research, as a nerdy academic tends to do, and you were telling me you were
[02:46.500 --> 02:49.260]  studying several different policing topics.
[02:49.260 --> 02:50.940]  What drew you to the issue of de-policing?
[02:50.940 --> 02:51.940]  Yeah.
[02:52.460 --> 02:57.700]  So de-policing has been something that has come to kind of the forefront of public attention.
[02:57.700 --> 03:02.100]  I mean, we saw with Michael Brown and Ferguson in 2014, there was a lot of conversations
[03:02.100 --> 03:08.460]  about this Ferguson effect, whether police were pulling back in response to public criticism.
[03:08.460 --> 03:13.020]  And we saw this topic really renewed in 2020 in ways that we hadn't seen in a really long
[03:13.020 --> 03:14.020]  time.
[03:14.020 --> 03:19.420]  I mean, at the time, there were calls to defund and even abolish the police across the nation.
[03:19.900 --> 03:23.740]  And one of the key questions, as you mentioned, that kept popping up in relation to this narrative
[03:23.740 --> 03:26.300]  is what happens when the police pull back?
[03:26.300 --> 03:31.100]  If police do pull back, is that going to have any kind of impact on crime?
[03:31.100 --> 03:36.540]  And we haven't seen a ton of research on the impact of these standard policing practices
[03:36.540 --> 03:38.100]  in general.
[03:38.100 --> 03:42.900]  We see evaluations of really specific strategies, things like hot spot policing and directed
[03:42.900 --> 03:43.900]  high crime areas.
[03:43.900 --> 03:47.300]  We see evaluations on things like focus deterrence.
[03:47.300 --> 03:50.860]  But we just haven't seen as much about these kind of general police practices.
[03:50.860 --> 03:51.860]  Okay.
[03:51.860 --> 03:55.960]  Now, during the intro, I hope I didn't steal any of your thunder.
[03:55.960 --> 04:00.540]  But from a broader perspective, why is research into de-policing of value or relevance to
[04:00.540 --> 04:02.660]  the police or even the public?
[04:02.660 --> 04:06.940]  And I do actually want to ask, de-policing sounds a little awkward and clunky.
[04:06.940 --> 04:08.620]  Is that what we're saying?
[04:08.620 --> 04:12.220]  I mean, we refer to it, I think, in the paper that we're discussing as police pullbacks,
[04:12.220 --> 04:15.540]  but it seems like de-policing is kind of the standard term.
[04:15.540 --> 04:20.220]  So it's called different things, but I agree with you.
[04:20.220 --> 04:24.380]  So what's the value from that, again, from that broader perspective, what's the relevance
[04:24.380 --> 04:27.260]  to policing or the public for understanding this?
[04:27.260 --> 04:31.300]  I think for the police and the public, we just need to know if what the police do impacts
[04:31.300 --> 04:32.300]  crime.
[04:32.300 --> 04:36.780]  We know that policing has been described as the hodgepodge of responsibilities.
[04:36.780 --> 04:39.300]  Officers do a ton of different things during the course of their careers.
[04:39.300 --> 04:42.460]  They do a ton of different things during the course of their shifts.
[04:42.460 --> 04:46.580]  We just don't have a good understanding of what types of those activities have crime
[04:46.580 --> 04:50.940]  reduction benefits and what types may not have those same effects.
[04:50.940 --> 04:54.580]  And so for both the police and the public, we need to understand if we're asking officers
[04:54.580 --> 04:59.620]  to do these things, whether that's pedestrian stops or vehicle stops, is that going to reduce
[04:59.620 --> 05:00.620]  crime?
[05:00.620 --> 05:02.100]  Is that going to make the community safer?
[05:02.100 --> 05:05.860]  And by answering those questions, we can really guide police resources in a more efficient
[05:05.860 --> 05:06.860]  way.
[05:06.860 --> 05:07.860]  Oh, very good.
[05:07.860 --> 05:12.140]  Now, I gave the briefest of explanations for de-policing in the intro.
[05:12.220 --> 05:16.820]  Can you expand on why COVID, particularly, and then George Floyd, that incident caused
[05:16.820 --> 05:21.140]  the police officers to reduce their proactive, and we're talking mostly proactive, aren't
[05:21.140 --> 05:22.140]  we?
[05:22.140 --> 05:23.140]  Yes.
[05:23.140 --> 05:24.140]  Mostly proactive activities.
[05:24.140 --> 05:27.740]  We know that officers didn't stop showing up if someone called 911 or if there was a
[05:27.740 --> 05:29.620]  major event in progress.
[05:29.620 --> 05:33.100]  This is more, you know, those things that an officer would self-initiate.
[05:33.100 --> 05:38.020]  So seeing something suspicious, engaging in potentially a discretionary arrest, a vehicle
[05:38.500 --> 05:43.020]  for something that was, you know, not a major violation, a pedestrian stop, those types
[05:43.020 --> 05:44.020]  of things.
[05:44.020 --> 05:45.020]  Okay.
[05:45.020 --> 05:46.020]  Now, okay.
[05:46.020 --> 05:52.100]  So you studied Denver, which is, you know, a large, a fairly large city, but why only
[05:52.100 --> 05:53.100]  one city?
[05:53.100 --> 05:55.260]  Would it have been better to compare a few different cities?
[05:55.260 --> 05:56.260]  Yes.
[05:56.260 --> 05:57.260]  Yeah.
[05:57.260 --> 06:00.660]  I think I want to go back to your previous question about why we think de-policing might
[06:00.660 --> 06:02.740]  have occurred in response to COVID and George Floyd.
[06:02.740 --> 06:03.740]  Just really briefly.
[06:03.740 --> 06:04.740]  Sure.
[06:04.740 --> 06:05.740]  Fine.
[06:05.740 --> 06:06.740]  Yeah.
[06:07.220 --> 06:09.260]  Of course, just people stayed home, right?
[06:09.260 --> 06:11.900]  There were stay at home orders across the country.
[06:11.900 --> 06:13.900]  People weren't really leaving their houses as much.
[06:13.900 --> 06:18.420]  So there just wasn't as much opportunity for police to engage in that type of proactive
[06:18.420 --> 06:19.420]  activity.
[06:19.420 --> 06:20.580]  We didn't have cars on the street.
[06:20.580 --> 06:22.440]  People were not walking around.
[06:22.440 --> 06:26.180]  So that's one reason we might have seen a police pullback during that time period.
[06:26.180 --> 06:32.100]  We also saw many agencies implement restrictions on the types of traffic offenses and stops
[06:32.100 --> 06:33.100]  to conduct.
[06:33.100 --> 06:35.980]  And this was really just to protect officer health and safety, right?
[06:36.100 --> 06:38.540]  We didn't know a lot about COVID in the beginning.
[06:38.540 --> 06:42.620]  And so we wanted to make sure we weren't putting people at risk of getting infected for things
[06:42.620 --> 06:43.620]  that were really minor.
[06:43.620 --> 06:47.840]  So there was kind of this policy implication of, you know, we don't want you to get sick.
[06:47.840 --> 06:50.660]  So maybe don't engage in this proactive behavior.
[06:50.660 --> 06:53.460]  And then we did see a lot of officers get sick during COVID.
[06:53.460 --> 06:56.740]  And that resulted in staffing shortages in a lot of places.
[06:56.740 --> 07:00.740]  And so when you have these staffing shortages, officers just don't have as much time to engage
[07:00.740 --> 07:02.380]  in that proactive stuff.
[07:02.380 --> 07:05.900]  They can't just conduct traffic stops and pedestrian stops when they're going from
[07:05.900 --> 07:07.520]  call to call to call.
[07:07.520 --> 07:12.540]  So for COVID, I think there's kind of very specific reasons we might have seen de-policing.
[07:12.540 --> 07:14.680]  For George Floyd, it's obviously different.
[07:14.680 --> 07:16.860]  We had this major incident in Minneapolis.
[07:16.860 --> 07:22.300]  We had protests across the nation and across the world asking for less policing.
[07:22.300 --> 07:23.900]  And officers heard that.
[07:23.900 --> 07:25.300]  They saw that.
[07:25.300 --> 07:29.540]  And one of the questions that we had was, did that result in them pulling back and saying,
[07:29.540 --> 07:35.740]  OK, you know, if this is what the conversation is, sure, we won't do this anymore.
[07:35.740 --> 07:38.380]  I'm concerned that I might be involved in a critical incident.
[07:38.380 --> 07:43.580]  I think there are a lot of reasons officers could be hesitant to be proactive in those
[07:43.580 --> 07:45.380]  times of turmoil.
[07:45.380 --> 07:50.380]  I think other research beyond ours would need to figure out what that exact reason is.
[07:50.380 --> 07:56.260]  But we certainly have this national narrative about police pulling back in response to these
[07:56.260 --> 07:57.260]  protests.
[07:57.260 --> 07:58.260]  Right.
[07:58.260 --> 08:01.140]  I remember reading, I live in Western New York, and I was reading a policy of a local
[08:02.140 --> 08:07.820]  about four or five years ago, and it had changed specifically because of COVID that they were
[08:07.820 --> 08:13.100]  to, for most misdemeanors, don't quote me on this, most misdemeanors, it was issuing
[08:13.100 --> 08:14.100]  an appearance ticket.
[08:14.100 --> 08:16.500]  The exception would have been domestic violence.
[08:16.500 --> 08:22.060]  But also getting back then to Denver, why Denver?
[08:22.060 --> 08:23.580]  Why not other cities?
[08:23.580 --> 08:27.620]  And I think another question I have about Denver is, how did you get into, any large
[08:27.620 --> 08:32.060]  city is going to, you know, often be resistant to letting researchers in the door.
[08:32.060 --> 08:33.060]  We all know that.
[08:33.060 --> 08:34.060]  So how did you get in there too?
[08:34.060 --> 08:35.060]  Sure.
[08:35.060 --> 08:36.060]  Yeah, that's a great question.
[08:36.060 --> 08:39.860]  So the reason that we went with Denver is Denver actually made it very easy for us.
[08:39.860 --> 08:42.420]  All of the data that we use is publicly available.
[08:42.420 --> 08:44.540]  So it's on their open data portal.
[08:44.540 --> 08:45.540]  It's on their website.
[08:45.540 --> 08:46.540]  You can go there right now.
[08:46.540 --> 08:47.620]  You could download.
[08:47.620 --> 08:49.260]  We needed really specific data.
[08:49.260 --> 08:53.620]  So they, because we're interested in crime and police activity, we needed data about
[08:53.620 --> 08:55.020]  crime and police activity.
[08:55.700 --> 09:00.380]  And we also know that those things vary across time and they vary across places and neighborhoods.
[09:00.380 --> 09:06.180]  And so we needed this data to have both a date, a time and a location.
[09:06.180 --> 09:11.700]  So an XY coordinate where that particular crime or police stop occurred.
[09:11.700 --> 09:17.260]  All of that stuff publicly available on Denver's website, we downloaded data from 2016 to 2020.
[09:17.260 --> 09:22.380]  So that allowed us to get, you know, four years pre 2020 that we could have this kind
[09:22.380 --> 09:27.540]  of comparison period too, to see whether we did see changes in policing during COVID
[09:27.540 --> 09:29.580]  and George Floyd.
[09:29.580 --> 09:31.940]  So that was a huge, huge advantage of using Denver.
[09:31.940 --> 09:34.220]  Not everyone shares that same data.
[09:34.220 --> 09:38.820]  Obviously, it would be great to say, you know, we saw this in Denver and we also saw this
[09:38.820 --> 09:39.820]  in New York.
[09:39.820 --> 09:43.820]  We also saw this in LA and we saw this in, you know, Reno, Nevada, whatever it is.
[09:43.820 --> 09:46.500]  But not every agency has that.
[09:46.500 --> 09:48.900]  So you didn't even have to play nice and ask permission.
[09:48.900 --> 09:49.900]  It's just there.
[09:49.900 --> 09:52.020]  No, it was just there for us.
[09:52.020 --> 09:53.340]  Which we greatly appreciate it.
[09:53.340 --> 09:57.180]  So thank you to the city of Denver for making our lives easy.
[09:57.180 --> 10:01.140]  It's like Texas, Dallas, Houston, San Antonio, a lot of their data.
[10:01.140 --> 10:04.980]  I know from the use of force research I've done, their data is just right there online.
[10:04.980 --> 10:05.980]  It's online.
[10:05.980 --> 10:06.980]  You can just download it and it's available.
[10:06.980 --> 10:10.580]  It might not be nearly as rich as if you have somebody inside, but okay.
[10:10.580 --> 10:11.580]  So I digress.
[10:11.580 --> 10:16.300]  They also had like, so you had different neighborhoods available in Denver then?
[10:16.300 --> 10:17.300]  Yes.
[10:17.580 --> 10:19.820]  Yes, yeah, Denver is a very neighborhood based city.
[10:19.820 --> 10:22.300]  So it has 78 unique neighborhoods.
[10:22.300 --> 10:25.500]  If you ask people who are familiar with Denver about the neighborhoods, they know about the
[10:25.500 --> 10:26.500]  neighborhoods.
[10:26.500 --> 10:29.180]  You could say the names of certain neighborhoods and they'll tell you whether that's a good
[10:29.180 --> 10:31.660]  or a bad neighborhood.
[10:31.660 --> 10:36.140]  And you know, for anyone who stays or does policing, we know that these things differ
[10:36.140 --> 10:37.340]  across place, right?
[10:37.340 --> 10:42.180]  Like not every single location has the same exposure to crime, not every location has
[10:42.180 --> 10:43.940]  the same exposure to police activity.
[10:44.580 --> 10:48.700]  And so the neighborhoods feature of Denver was also really helpful for us.
[10:48.700 --> 10:52.500]  Most of the time when you want to look at neighborhoods in other cities, they don't
[10:52.500 --> 10:54.820]  have really clear boundaries.
[10:54.820 --> 10:59.660]  Like it might be, you know, you think it's from the river to this particular street is
[10:59.660 --> 11:06.820]  one neighborhood and this is another neighborhood, but there's no easy like definition or distinction.
[11:06.820 --> 11:07.940]  Denver doesn't have that problem.
[11:07.940 --> 11:10.780]  Those neighborhoods are very clear, very distinct.
[11:10.780 --> 11:15.220]  And then I guess I will say one other benefit of using a single city over using multiple
[11:15.220 --> 11:20.700]  is that, you know, offense codes vary across cities, police policies vary across cities.
[11:20.700 --> 11:24.180]  So, you know, when you use just one city, of course, you can only say things about that
[11:24.180 --> 11:27.180]  one city, but you're also not comparing apples to oranges.
[11:27.180 --> 11:31.780]  Like I know what a drug arrest is in Denver or a disorder arrest is in Denver.
[11:31.780 --> 11:36.260]  And I don't have to worry about whether that's comparable to another place.
[11:36.260 --> 11:37.260]  Got it.
[11:37.260 --> 11:38.260]  Okay.
[11:38.420 --> 11:43.020]  You wanted to find out if the police were backing off their activity and if it had an
[11:43.020 --> 11:44.020]  impact on crime.
[11:44.020 --> 11:46.220]  So tell us how the study was done.
[11:46.220 --> 11:47.220]  Yeah.
[11:47.220 --> 11:50.300]  So the first steps was just downloading this massive amount of data, right?
[11:50.300 --> 11:54.960]  So we had to go out, we had to get our, you know, our crime data or police activity data.
[11:54.960 --> 12:00.180]  Because we know that neighborhoods matter, we went to the American census, we got community
[12:00.180 --> 12:04.420]  data from the census website, also publicly available.
[12:04.420 --> 12:07.660]  Crime policing also depend on things like precipitation.
[12:07.660 --> 12:12.620]  So we even went to the Environmental Protection Agency's website and got weekly measures
[12:12.620 --> 12:18.100]  of inches of rain and snow, because we thought that might make a difference.
[12:18.100 --> 12:20.980]  And then really, because we're looking at 2020, which was weird, right?
[12:20.980 --> 12:25.420]  We had stay at home orders, we had, you know, people not traveling as much, not going out
[12:25.420 --> 12:26.420]  as much.
[12:26.420 --> 12:30.940]  We also got restaurant reservation data from Open Table, so that we could look at things
[12:30.940 --> 12:32.740]  like population mobility.
[12:32.740 --> 12:39.860]  So we really wanted to know, you know, policing crime varies across a ton of different conditions.
[12:39.860 --> 12:44.020]  We wanted to first just make sure that we were able to account for all of these other
[12:44.020 --> 12:49.020]  things before we started looking at our main research questions, which of course were,
[12:49.020 --> 12:51.300]  did policing decline after COVID?
[12:51.300 --> 12:53.740]  Did policing decline after George Floyd?
[12:53.740 --> 12:59.520]  And then did those pullbacks in policing have any impact on violent and property crime?
[12:59.520 --> 13:03.360]  So that was really our first step, was just pulling together this big, massive data set.
[13:03.360 --> 13:07.440]  Again, we were lucky to work with Denver, where all of that data was publicly available.
[13:07.440 --> 13:12.920]  Okay, so as other people who have worked with me know, statistics tends to hurt my brain.
[13:12.920 --> 13:20.240]  So when it comes to what you found, did officers change their behavior, and did it impact crime?
[13:20.240 --> 13:21.240]  Yes.
[13:21.240 --> 13:22.240]  Yeah, I'll keep it short.
[13:22.240 --> 13:24.440]  I'll try to keep it high level.
[13:24.440 --> 13:25.440]  So first, yes.
[13:25.440 --> 13:26.440]  Oh, no, that's fine.
[13:26.640 --> 13:27.640]  That's fine.
[13:27.640 --> 13:32.080]  If you want to dig into the dirt on this one, I think the listeners will be very interested
[13:32.080 --> 13:33.720]  to find out what you know.
[13:33.720 --> 13:34.720]  Okay.
[13:34.720 --> 13:35.720]  Yeah.
[13:35.720 --> 13:39.040]  So first, we just wanted to know if discretionary policing changed.
[13:39.040 --> 13:42.660]  And to do that, we wanted to look at four different discretionary outcomes.
[13:42.660 --> 13:47.660]  So pedestrian stops, vehicle stops, drug arrests, and disorder arrests.
[13:47.660 --> 13:52.840]  So these are kind of the things that we figured, you know, police might have more individual
[13:52.840 --> 13:55.200]  ability to say, I'm going to do this, or I'm not going to do this.
[13:55.200 --> 13:57.600]  Things that wouldn't be affected.
[13:57.600 --> 14:01.840]  Like you're still going to go out and make an arrest for an aggravated assault or a homicide.
[14:01.840 --> 14:07.560]  So first and foremost, we did find, yes, police activity did reduce after COVID, and it did
[14:07.560 --> 14:10.400]  decline after George Floyd's murder.
[14:10.400 --> 14:14.280]  So I'll give you some just rough percentages so you have a sense of what this looks like
[14:14.280 --> 14:15.280]  in Denver.
[14:15.280 --> 14:16.280]  Sure.
[14:16.280 --> 14:20.480]  Between COVID and George Floyd, so that first stay at home order issued in the state of
[14:20.480 --> 14:21.480]  Denver.
[14:21.480 --> 14:27.160]  There was a 50% reduction in pedestrian stops, a 40% reduction in vehicle stops, 72% fewer
[14:27.160 --> 14:31.840]  drug arrests, and 27% fewer disorder arrests in the city of Denver.
[14:31.840 --> 14:37.000]  So pretty substantial notable changes in police activity.
[14:37.000 --> 14:40.520]  If we look at what happened after the death of George Floyd and throughout the remainder
[14:40.520 --> 14:44.720]  of 2020, we continue to see this lower level of police activity.
[14:44.720 --> 14:52.400]  61% fewer pedestrian stops, 48% fewer vehicle stops, 74% fewer drug arrests, and 34% fewer
[14:52.400 --> 14:53.520]  disorder arrests.
[14:53.520 --> 14:56.680]  So really notable reductions there.
[14:56.680 --> 15:01.160]  It's one of those things I'm thinking to myself as you're talking that I could understand
[15:01.160 --> 15:06.120]  the disorder arrests because I would think because maybe less people are out doing things
[15:06.120 --> 15:10.320]  in a public way, but there might be more disorder in a house.
[15:10.320 --> 15:17.120]  But the drug arrests, 70 something percent seems like an enormous amount.
[15:17.120 --> 15:20.680]  Now you also mentioned the idea of setting those different neighborhoods and when I was
[15:20.680 --> 15:25.040]  reading the article, you had mentioned what are called disadvantaged neighborhoods, poor
[15:25.040 --> 15:27.200]  places.
[15:27.200 --> 15:30.240]  Any particular results in those neighborhoods as well?
[15:30.240 --> 15:32.080]  Places where crime normally does occur?
[15:32.080 --> 15:33.080]  Yes.
[15:33.080 --> 15:34.080]  Yeah.
[15:34.080 --> 15:37.080]  We really expected when we started this study that these police pullbacks would have the
[15:37.080 --> 15:41.000]  most impact in these disadvantaged neighborhoods where crime is occurring.
[15:41.000 --> 15:45.520]  So they would be left unprotected and they would have higher crime spikes than your more
[15:45.520 --> 15:48.360]  standard or more advantaged neighborhoods.
[15:48.360 --> 15:53.160]  And really what we saw was limited solely to property offenses.
[15:53.160 --> 15:58.200]  So when police pulled back into disadvantaged neighborhoods, particularly when they conducted
[15:58.200 --> 16:03.680]  lower levels of pedestrian stops, we did see kind of increases in property offenses in
[16:03.680 --> 16:05.640]  those neighborhoods.
[16:05.680 --> 16:10.480]  And so really for us, what we take this as is these neighborhoods likely have high levels
[16:10.480 --> 16:12.600]  of police activity in general.
[16:12.600 --> 16:17.720]  So it might take really, really big reductions for them to feel these huge changes in crime
[16:17.720 --> 16:19.320]  as a result.
[16:19.320 --> 16:23.160]  But this does matter and it shows that the specific type of discretionary police activity
[16:23.160 --> 16:26.000]  matters where pedestrian stops are making a difference.
[16:26.000 --> 16:31.200]  We didn't see that same effect for either the drug or disorder arrests or vehicle stops.
[16:31.200 --> 16:35.440]  So we really do think kind of teasing out this nuance can be helpful for both the public
[16:35.440 --> 16:40.480]  and the police understanding like what types of policing we need, particularly in these
[16:40.480 --> 16:45.000]  neighborhoods where we have these conversations about, you know, do they want the police and
[16:45.000 --> 16:47.240]  what do they want the police to be doing?
[16:47.240 --> 16:48.240]  Right.
[16:48.240 --> 16:51.920]  Well, those are the kinds of comments after George Floyd, like not just defund, but eliminate
[16:51.920 --> 16:52.920]  the police.
[16:52.920 --> 16:57.160]  And now what we're seeing here is if you don't do any of that, but the police still naturally
[16:57.160 --> 17:02.440]  just back off, whether it's because of COVID or just they don't want to get themselves
[17:02.440 --> 17:07.520]  into trouble, that there's not necessarily going to be a good outcome from that.
[17:07.520 --> 17:08.520]  Yes.
[17:08.520 --> 17:09.520]  Yeah, exactly.
[17:09.520 --> 17:13.480]  So that was kind of the second part of our study was first, just did this deep policing
[17:13.480 --> 17:14.480]  occur?
[17:14.480 --> 17:17.960]  And the second part is how did that impact crime or did it?
[17:17.960 --> 17:18.960]  And we did find yes.
[17:18.960 --> 17:23.400]  I mean, particularly for pedestrian stops, when there were these reductions in pedestrian
[17:23.400 --> 17:28.620]  stops in neighborhoods in Denver, violent crime went up and property crime went up.
[17:28.620 --> 17:31.880]  So pedestrian stops really seem to be driving this.
[17:31.880 --> 17:36.200]  We didn't see those same effects for vehicle stops, disorder arrests or drug arrests.
[17:36.200 --> 17:43.240]  So really, it's limited to kind of that pedestrian stop finding and, you know, it kind of shows
[17:43.240 --> 17:47.800]  that we can change what police are doing without these major impacts on crime that we might
[17:47.800 --> 17:48.800]  expect.
[17:48.800 --> 17:49.800]  Yeah.
[17:49.800 --> 17:54.640]  When you say that pedestrian stops, well, reduction in pedestrian stops led to more
[17:54.640 --> 17:55.640]  crime.
[17:55.640 --> 18:01.800]  It's, I'm reflecting back on the idea of hotspot policing where you have done the
[18:01.800 --> 18:05.840]  necessary stops, but when officers are in the hotspots engaging the public, those aren't
[18:05.840 --> 18:09.480]  necessarily stops, but they're still engaging the public in a different, hopefully, you
[18:09.480 --> 18:15.400]  know, plus positive way that that then contributes according to what the research tells us so
[18:15.400 --> 18:19.160]  far that contributes to a reduction in crime in those neighborhoods.
[18:19.160 --> 18:27.320]  So your study seems to piggyback in a way to those other studies about hotspots.
[18:27.320 --> 18:29.280]  Yeah, absolutely.
[18:30.280 --> 18:35.720]  Even though we didn't look at specific changes in strategy or policy or agency directions
[18:35.720 --> 18:39.960]  to conduct these things in different places, we did see that those changes matter.
[18:39.960 --> 18:45.160]  So officers, just using their own discretion, pulling back did have a crime impact, particularly
[18:45.160 --> 18:46.440]  in these places.
[18:46.440 --> 18:51.000]  And I think, you're right, it does align with this hotspots idea of we should put police
[18:51.000 --> 18:54.400]  where they're needed and give them that direction, right?
[18:54.400 --> 18:57.920]  And so we're not just blanketing a whole city with this.
[18:57.920 --> 19:02.600]  Anything else I wanted to ask you about that you found in your article mentions, I'm going
[19:02.600 --> 19:09.200]  to read the quote that discretionary policing behavior has hit, I'm sorry, I'll start again.
[19:09.200 --> 19:14.040]  Discretionary policing behaviors had slowly begun to return towards normal levels in May.
[19:14.040 --> 19:18.240]  And when I read that, I thought to myself, you know, this was a little bit, just a little
[19:18.240 --> 19:21.000]  bit before there was the George Floyd incident.
[19:21.000 --> 19:25.760]  And I found it interesting that de-policing only lasted, at least, you know, from a COVID
[19:25.840 --> 19:30.240]  perspective for a few months, there seemed to be a return to the normal, I want to do
[19:30.240 --> 19:31.920]  these kinds of things from the police.
[19:32.560 --> 19:33.480]  You have any thoughts on that?
[19:34.240 --> 19:38.680]  Yeah. So what we did is we, you know, as I mentioned, we had data going back to 2016.
[19:38.800 --> 19:42.920]  And so we created kind of this three week rolling average of police activity.
[19:42.920 --> 19:49.720]  So for the same three week period from 2019, 2018, 2017, 2016, we were able to see what
[19:49.720 --> 19:55.080]  was normal for levels of policing in Denver for those years and compare it to that 2020
[19:55.080 --> 19:59.720]  policing. And so I know that like the normal amount of policing seems kind of funky,
[19:59.720 --> 20:04.280]  right? But we were able to look at that trend over time, trend across neighborhoods and
[20:04.280 --> 20:06.920]  see, you know, where was this before this event?
[20:07.160 --> 20:11.640]  And we did see it getting closer and closer to the levels it was at in those prior years.
[20:11.840 --> 20:15.440]  And I think you're right. I mean, everyone after COVID talked about this return to normal.
[20:15.440 --> 20:19.200]  And I don't know that we ever fully have returned to normal or what normal particularly
[20:19.200 --> 20:23.920]  is. But it did seem like after those kind of first few months of COVID with the heaviest
[20:23.920 --> 20:28.840]  shutdowns, people were interested in getting back out and doing things again and kind
[20:28.840 --> 20:32.080]  of going back into their social and work routines.
[20:32.080 --> 20:35.600]  And so, yeah, we kind of did see that trend in police activity as well.
[20:35.720 --> 20:38.280]  Climbing back up. I just I just thought that was interesting.
[20:39.080 --> 20:43.800]  A return to return to the normal part of a police officer that and again, without this
[20:43.800 --> 20:47.600]  information, you can't you can't ask the question of, OK, who was the officer with the
[20:47.600 --> 20:49.920]  older officers, younger officers, you know, those kinds of things.
[20:50.920 --> 20:56.440]  A second question I had as I read the paper, I thought it was interesting that some
[20:56.440 --> 21:01.800]  proactive street level activity might be avoided because of the social environment with
[21:01.800 --> 21:05.120]  regard to to to to the, you know, George Floyd.
[21:05.240 --> 21:09.480]  The social environment was we don't like cops, at least, you know, that it came across
[21:09.480 --> 21:15.800]  that way. But some some policing might be reduced through the policy of the agency.
[21:16.360 --> 21:20.000]  But do you have any thoughts on that approach or that that way of thinking, the difference
[21:20.000 --> 21:22.160]  between a social environment versus a policy?
[21:23.120 --> 21:24.520]  Yeah, no, I think you're absolutely right.
[21:24.520 --> 21:28.720]  It's interesting that there is this difference in policy, specifically, like you said,
[21:28.720 --> 21:30.600]  would focus on things like misdemeanors.
[21:30.600 --> 21:32.960]  It would focus on things like traffic stops in certain places.
[21:32.960 --> 21:35.080]  So for COVID, you could see that being really clear.
[21:36.000 --> 21:37.400]  For Floyd, it's harder, right?
[21:37.400 --> 21:42.200]  Like, we don't necessarily know what it is people were asking for less of.
[21:42.280 --> 21:46.280]  And so individual officers on the street might have interpreted that very differently and
[21:46.280 --> 21:51.160]  said, you know, I'm not going to engage in a drug arrest or a disorder arrest or stop
[21:51.160 --> 21:53.680]  someone on the street because I don't think they want that.
[21:53.680 --> 21:57.360]  But another officer might have said, you know, I think they actually do want me to do this
[21:57.360 --> 22:00.120]  because that's what keeps these neighborhoods safe.
[22:00.280 --> 22:04.160]  So I think there's kind of this subjective judgment on the value of the officer.
[22:04.160 --> 22:08.320]  And as you just mentioned, without knowing who the individual officers were or their
[22:08.320 --> 22:11.840]  background characteristics, we really couldn't answer that with this question.
[22:11.840 --> 22:14.560]  But I do think that, you know, it's interesting.
[22:14.560 --> 22:18.360]  There are things we can change through policy and think that are caused by policy and
[22:18.360 --> 22:23.360]  others that are purely decisions officers make in their in their daily activity.
[22:23.600 --> 22:28.880]  Yeah. OK. So, OK, overall, what are a couple of the implications for the police
[22:28.880 --> 22:34.040]  practice or itself or police personnel or just police leaders when it comes to what
[22:34.040 --> 22:35.040]  your research tells us?
[22:36.560 --> 22:39.920]  Sure. I think the first implication is we really just need to pay attention to how
[22:39.920 --> 22:44.640]  much proactivity officers are engaging and the downstream impact of that
[22:44.840 --> 22:47.960]  activity. Right. So how many stops are you conducting?
[22:48.360 --> 22:49.760]  Are those stops successful?
[22:50.720 --> 22:54.160]  You know, we found that officers made a huge reduction in traffic stops in Denver,
[22:55.360 --> 22:59.280]  but we didn't find any relationship between those stops and violent crime or property
[22:59.280 --> 23:03.240]  crime. That's not to say that traffic stops aren't important for like traffic safety
[23:03.240 --> 23:07.600]  and other reasons. But if we're using traffic stops as a crime prevention strategy, I
[23:07.600 --> 23:12.000]  think our findings suggest that you can make those reductions, which would likely
[23:12.000 --> 23:16.800]  satisfy segments of your community without having these huge negative impacts on
[23:16.800 --> 23:22.000]  crime. So first, I think it's just paying attention to what those patterns look
[23:22.000 --> 23:23.440]  like and what the outcomes are.
[23:24.360 --> 23:26.120]  And that will help you focus your resources.
[23:26.120 --> 23:30.000]  So if you do want to focus on traffic safety, roadway safety, that's great.
[23:30.640 --> 23:31.880]  That's something that makes sense.
[23:31.880 --> 23:36.560]  If you're more focused on using traffic stops as a violent property crime reduction,
[23:36.560 --> 23:38.920]  it's probably not going to help you as much.
[23:38.960 --> 23:44.640]  OK. You know, the second thing I would say is we looked at really just this general
[23:44.640 --> 23:49.480]  proactive police activity and we did find these effects, right, of officers pulling
[23:49.480 --> 23:54.920]  back and that pullback having an association with violent and property crime.
[23:56.600 --> 24:01.560]  I think what that means for agencies is while we looked at this generally, what we
[24:01.560 --> 24:03.920]  should be doing is targeting things specifically.
[24:04.000 --> 24:08.280]  So figuring out where this will have the most impact, figuring out where we can help
[24:08.280 --> 24:14.000]  communities the most and surgically target our proactive policing efforts there so
[24:14.000 --> 24:18.960]  that we're not kind of putting officers in places where they don't necessarily need
[24:18.960 --> 24:20.920]  to be, where it's not necessarily going to make a difference.
[24:21.880 --> 24:25.400]  So I think it's it's kind of a question of sitting down and thinking strategically
[24:25.400 --> 24:29.120]  about what the goal of the agency is and how you can align your resources with that
[24:29.120 --> 24:32.680]  goal. OK, so that would be kind of my second.
[24:32.840 --> 24:37.080]  I know strategic planning is like not people's favorite thing to do, but I do think
[24:37.080 --> 24:41.480]  it's helpful. And then again, I think focusing on different types of neighborhoods is
[24:41.480 --> 24:45.360]  helpful, too. So it's not what we're finding in these disadvantaged neighborhoods,
[24:45.360 --> 24:47.880]  right, is that policing pulling back is making a difference.
[24:47.880 --> 24:49.920]  So we need to police those neighborhoods.
[24:50.440 --> 24:51.960]  And there is this challenge, right.
[24:51.960 --> 24:56.120]  Oftentimes these are high crime communities that experience a lot of policing, and
[24:56.120 --> 24:59.080]  sometimes they're the ones who are saying, you know, we've had enough.
[24:59.920 --> 25:03.760]  Really, I think what this is showing is that we don't need less policing in these
[25:03.760 --> 25:07.720]  places. We need kind of better policing that really matches the needs of those
[25:07.720 --> 25:12.960]  areas. And so focusing on kind of the neighborhood level approaches as opposed to
[25:12.960 --> 25:17.080]  looking at something citywide, I think can be really useful for police leaders who
[25:17.080 --> 25:20.680]  want to improve their services, particularly in those types of areas.
[25:21.360 --> 25:22.560]  Those would be my takeaways.
[25:22.880 --> 25:27.160]  Well, very good, Jessica or Jesse, I really do appreciate your time.
[25:27.640 --> 25:29.480]  This was fascinating research.
[25:29.720 --> 25:32.240]  I think it'll be very helpful for people to listen to.
[25:32.240 --> 25:37.440]  I'll make sure that I post the journal title and where it can be found in the in
[25:37.440 --> 25:42.400]  the show notes. And if anybody wants to track you down, you're at the University
[25:42.400 --> 25:43.840]  of Cincinnati. Everybody can find you.
[25:43.840 --> 25:50.400]  I did. And if people want to contact me, the emails in the show notes, I can give
[25:50.400 --> 25:52.720]  you more information on where to get the article.
[25:53.640 --> 25:55.800]  Dr. Huff, thank you so very much for your time.
[25:55.800 --> 25:57.120]  I really do appreciate it.
[25:57.680 --> 25:58.720]  Yeah. Thank you so much.
[25:58.840 --> 25:59.600]  All right. Have a great day.
[26:00.000 --> 26:00.600]  All right. You too.
[26:04.800 --> 26:08.560]  That's it for this episode of the Police in Service Training podcast.
[26:09.000 --> 26:12.320]  I want to thank you, the listener, for spending your valuable time here.
[26:12.680 --> 26:16.040]  If you like what you've heard, please tell a friend to subscribe on Apple
[26:16.040 --> 26:18.840]  podcast or wherever they get their podcasts.
[26:19.120 --> 26:21.120]  And please take a moment to review this podcast.
[26:21.600 --> 26:25.760]  If you have any questions or comments, positive or negative, or if you think
[26:25.880 --> 26:30.000]  I should cover a specific topic, feel free to send me an email, which you
[26:30.000 --> 26:33.960]  can find in the show notes, or you can find me on bluesky using the handle
[26:34.280 --> 26:38.920]  at policeinservice.bsky.social.
[26:39.440 --> 26:40.880]  Thanks very much and have a great day.