DC has made meaningful improvements in safety but there’s still an unacceptable level of violence in our city. Shootings continue to needlessly end lives, maim survivors, terrify bystanders and poison DC’s image on social and traditional media. While the Metropolitan Police Department (MPD) can point to legitimately impressive statistics like “The number of people shot is down 35%” there is still more that DC and the Federal agencies responsible for so much of our criminal justice system should be doing. This post will cover what we know about DC’s gun violence problem and some apparent gaps in our approach.
“Crimes with a gun” represent only about 7% of all reported crimes in DC but are 83% of homicides. Shootings specifically are an even smaller share of reported crime since many “violent crimes with a gun” are robberies where a suspect brandishes (but doesn’t shoot) a firearm. In 2023 there were 1,135 reported shootings, or 1 for every 598 resident Washingtonians. However our best evidence suggests that many shootings are never reported to police since there were 8,690 ShotSpotter alerts (i.e. detected gunshots) in 2023; which is 1 alert for every 78 Washingtonians. The vast majority of reported shootings and ShotSpotter alerts occur in DC’s poorer neighborhoods in police districts 6D and 7D:
*Note that 7 of the 1,135 reported shootings weren’t mapped to a police district*
These numbers help set the critical context that DC’s gun violence problem is driven by a very small number of people. The vast majority of Washingtonians in every neighborhood are not shooting at each other. This is a problem that a small, dangerous population imposes upon a populace that overwhelmingly wants the violence to stop.
One perceptual challenge is that even when the rate of violence decreases there’s still usually at least one reported shooting (and several ShotSpotter alerts or audible gunshots) every day. This real and constant drumbeat of shootings on social and traditional media has caused some people to believe that the reported decrease in gun violence this year is “fake.” However the fact is that all of DC’s gun violence indicators are moving in the same positive direction:
Reported gun homicides are down ~20% YTD (DC CrimeCards doesn’t have the 3 4/24/2024 gun deaths yet so this is an estimate)
Reported non-fatal shootings are down 35% YTD
The number of people shot is down 35% as of April 4th (with the number of juveniles shot down 59%)
Emergency Department Visits for Firearm Injury are down 30% YTD (note that the police do NOT control this data source)
In addition to these data points it’s worth noting that any coordinated effort to fake shooting data would be incredibly likely to leak to the media, the police union and/or Congressional Republicans. DC’s drop in gun violence so far this year is real and has saved ~10 lives.
But this statistical improvement is little comfort to the 43 families who have lost someone to gun violence so far this year. Or the many others who have lived through shootings or are simply sick of reading about violence on DC streets:
One of the most frustrating aspects of DC’s gun violence problem is that we know so much about the people doing the shooting and yet the criminal justice system has failed to stop them. DC has paid for two excellent “Gun Violence Problem Analysis” reports that identify the highest-risk 200-500 suspects who drive 60-70% of gun violence in DC. These named individuals are known to law enforcement and DC’s violence prevention programs. Without rehashing my previous post on this report here are the key points about this highest-risk population:
The average homicide suspect has been arrested 10 times for ~12 offenses at the time of the alleged murder.
Those 10 arrests only results in ~6.9 charged cases due to the United States Attorney’s Office (USAO) declining to prosecute
Of those 6.9 charged cases the USAO ends up dropping 3.6 without a conviction
While the average homicide suspect has 3.1 prior convictions, only 54% have any prior felony conviction. This suggests that a lot of these convictions are for misdemeanors; often the result of plea deals with the USAO.
Many of these individuals are involved in gangs or crews and engaging in shootings for incredibly stupid reasons: “There is a deadly mix of group/crew/gang members making music videos taunting or disrespecting their rivals that are posted on social media, and those videos spark or further inflame neighborhood conflicts that escalate into shootings”
90% of homicide suspects and 74% of non-fatal shooting suspects are adults. Juveniles are the majority of arrested carjacking suspects but DC’s problem with shootings is overwhelmingly among legal adults
Despite being well known to law enforcement and priorities for preventative outreach these men (over 90% of these suspects are male) continue to bounce around DC’s criminal justice system without being deterred, incapacitated or rehabilitated. The People of Promise program was intended to target this group for rehabilitation but the vast majority of eligible individuals either never enroll in the program or quickly disengage. It’s unclear what DC is doing to revamp the program but the Mayor’s proposed budget cuts the Office of Neighborhood Safety and Engagement (ONSE), the agency that oversees People of Promise and other violence prevention efforts, by 11%. The “Violence Intervention” programs specifically have a proposed 54% cut:
Money isn’t everything, but pulling back so dramatically on violence prevention puts even more pressure on enforcement to try and stem gun violence. We have mixed news on enforcement specific to gun violence. The good news is that MPD’s surge in arrests for violent crimes (+106% YTD) and gun possession (+58% YTD) continued in the latest March data:
This helps confirm that the large spike in the January-February data wasn’t a fluke and that MPD is truly making (and sustaining) vastly more arrests with less staffing. As a short aside, even the staffing data has some cause for optimism since MPD reports that they have received 47% more applicants YTD. The flurry of internal changes at MPD after Chief Smith took over included standing up a "new in-person recruiting unit.” MPD appears to be on pace to both grow and be more effective relative to the post-COVID years.
Given the large increase in arrests and overall falling reported gun violence it’s likely that MPD’s clearance rate for non-fatal shootings has increased from last year’s 26%. This is key because such a low clearance rate undermines deterrence and means that many shooters are out on the street to either shoot again or be targets themselves for retaliatory attacks. However there are still some best practices to solve more crimes where DC is lagging or it’s unclear what is going on due to a lack of transparency:
Cameras: MPD has expanded its network of cameras and has a plan to grow it further from ~350 (according to the MPD website) to 1,025 by 2028. Chief Smith described this in her budget hearing testimony: “The FY25 budget continues Mayor Bowser’s sustained commitment for expanding DC’s CCTV network on public space. We know that cameras support better investigations and prosecutions, and deter crime. In 2023 we deployed 48 new cameras, and will deploy 50 in 2024. The FY25 budget will accelerate this critical investment, supporting 200 more cameras. This investment will continue over the next four years until we reach 1,025 cameras in 2028. The FY25 budget also supports the necessary CCTV refresh program to ensure older cameras are updated so that images are clear and support investigations and prosecutions. They also help the city to deploy resources during major events, unrest, or threats. The FY25 budget funds the purchase of 47 new License Plate Readers, or LPRs, and critical system support. We use LPRs to help locate missing persons, carjacked or stolen vehicles, and vehicles used in violent crime.” The question here is if MPD could operationalize a faster rollout of this technology if the funds were frontloaded. This kind of technology is far more scalable than the laborious work of recruiting and training more officers. If MPD could deploy more cameras and LPRs faster it would be a worthwhile investment since it would increase the effectiveness of the ~$574M we intend to spend on MPD as a whole. These sorts of investments also benefit from now being embedded into a police department where the “Real Time Crime Center” is better able to use these tools to direct officers to apprehend suspects in the moment; not only to identify suspects after the fact.
Aerial Assets: MPD’s assessment that they could use almost 3X as many cameras as they have today helps illustrate the gaps in DC’s capabilities at a macro level. For a specific example, this thread on Twitter helps highlight how the suspect vehicle in a mass shooting was able to get away from police officers who were actively pursuing them in their police car. Even when the much misunderstood “chase policy” allows officers to pursue suspects the physics and danger of vehicular pursuits mean that even the highest-priority suspects can slip away. Building out a system of cameras, LPRs and preferably Unmanned aerial vehicles (UAVs) would make it much less likely that suspects could escape and that in turn would deter some shootings in the first place. Currently MPD has to rely on older, more expensive helicopters for aerial pursuits. This means that helicopter use has to be ruthlessly prioritized whereas cheaper and simply more available UAVs would mean that officers responding to reported shootings or ShotSpotter alerts could have an “eye in the sky” to help keep them safe and identify suspects. DC unfortunately sits in some of the most regulated airspace in America and it would take coordination with Federal agencies to allow MPD to use UAVs. DC at one point stood up a "Unmanned Aircraft System Working Group" but it doesn’t seem to have any posted meetings or information online. People tend to be very wary of “drones” and like other jurisdictions using UAVs (like Montgomery County) DC would need to develop regulations on their use. In general, it would seem reasonable to be able to send an unarmed UAV to any incident where today we send an armed agent of the state (i.e. a police officer). DC would need to avoid running afoul of the 2021 court ruling against Baltimore’s aerial surveillance program which was less targeted than “responding to a certain incident” and more like a constant camera in the sky. DC’s lack of public progress towards even piloting UAV technology is a real missed opportunity and if there is work going on MPD and the Bowser administration should announce it.
GPS: 17% of homicide suspects were under active Court Services and Offender Supervision Agency (CSOSA) supervision at the time of the alleged murder. However CSOSA appears to significantly under-utilize GPS or “electronic monitoring” for even its highest risk offenders. At the end of Fiscal Year 2023 CSOSA had 2,299 people in the “Maximum” supervision level and yet had only 382 (17%) on GPS monitoring. This is a mismatch between what CSOSA tells Congress about the risk level of these offenders (“Maximum”) and how they are failing to use all available tools to deter additional crimes.
Forensics: Since so many homicide suspects are people who “got away with” previous crimes we’ll never know how much of today’s gun violence problem stems from the disastrous closure of DC’s crime lab. Looking forward it is imperative that the Department of Forensic Sciences (DFS) process the enormous backlog of evidence that built up during the years the lab was unaccredited. DFS Director Diaz announced that some progress has been made: “To date in FY24, the FBU received Federal Bureau of Investigations (FBI) authorization for access to the Combined DNA Index System (CODIS) database, in which the unit has successfully reviewed and uploaded 140 cases, a process that is instrumental in identifying suspects and closing criminal cases in the District.” Unfortunately DFS earlier reported a backlog of 1,176 cases so depending on when DFS resumed CODIS uploads this implies that it could take over a year to work through the backlog unless they increase their pace. This sort of basic data transparency is unfortunately lacking at DFS which doesn’t maintain any timely operational data on their website. This near-complete lack of transparency makes DFS’ efforts to drive out their oversight body, the Science Advisory Board, even more concerning. As of today there are NO members on the Science Advisory Board. Proper oversight of the crime lab and fast-as-possible processing of the evidence backlog are critical to solve crimes and prevent another disastrous loss of accreditation.
These kinds of investments and improvements would likely help MPD solve more crimes across the board and shootings in particular. Increasing the certainty of punishment is key to deterring crime and giving prosecutors the chance to incapacitate offenders. Unfortunately we’ve already seen that DC’s prosecutors have missed chance after chance to do so with DC’s highest-risk suspects and recent data from the USAO doesn’t suggest much improvement.
Recently the USAO organized a triumphant press conference “to update the public on the process they put in place two years ago to review every firearms arrest in the District of Columbia, looking at data about the person arrested and data about the firearm they possessed to determine whether they should be prosecuted in U.S. District Court or in D.C. Superior Court.
The office and our partners refer to this effort as, “Project Safe Neighborhood” (PSN). In the two years since this PSN process was put in place, the office has accepted 127 firearms cases for federal prosecution.” US Attorney Matthew Graves claimed “The work we have done between this Project Safe Neighborhood effort and our other efforts have collectively taken hundreds of drivers of violence out of our community over the last couple years” and claimed “Removing this many drivers of violence has an impact on community safety.” Of course 127 cases over 2 years is barely a drop in the bucket compared to the scale of DC’s gun problem:
127 cases over 2 years is less than 6 per month. By contrast MPD made over 4,000 arrests for gun possession in 2022 and 2023. This means that “Project Safe Neighborhood” firearm possession cases are ~3% of the arrests that MPD brought to the USAO.
In 2022 and 2023 the USAO declined to prosecute 1,288 arrests for Carrying a Pistol Without a License (CPWL). This means that the USAO declined to prosecute at least 10 firearms cases for every 1 Project Safe Neighborhood case they bragged about to the media.
Relative to the 200-500 highest-risk “drivers of gun violence” these 127 Project Safe Neighborhood cases represent far less than even 1 case per individual
This kind of “look at our large number of prosecutions” without any context is typical of the USAO’s media operations. In fairness to the USAO they did cite their “other efforts” in addition to Project Safe Neighborhood but the extremely limited prosecution data they’ve shared shows only marginal improvement:
Since the Department of Justice (DOJ) announced a “surge” of resources to DC at the end of January, new USAO prosecutions in DC Superior Court have increased from 739 in January to 842 in March; a 14% increase.
However public/Congressional/media pressure without additional resources pushed the USAO to increase prosecutions from 421 per month in Fiscal Year (FY) 2022 to 739 by January 2024; a 76% increase. While this happened over a longer timeframe it’s still notable that pressuring the USAO drove a much larger increase in prosecution than this “surge” of resources.
Even the new “surge” level of prosecutions (842) is still less than the USAO’s output in September 2023 (860), when they were under intense pressure to close out FY 2023 with more favorable numbers.
This “surge” level of prosecutions is still 35% lower than the average monthly volume of new prosecutions (1,287) the USAO brought between FY 2014-2017.
Some people in DC’s criminal justice system have noted that the USAO has charged more cases in District Court and if so this is exactly the kind of data they should be releasing to the public. One police officer shared this feedback on the federal “surge”:
“The extra feds and extra [Assistant United States Attorneys] AUSAs are running into the same problem [as] MPD: DC Superior court judges and the existing DC USAO culture. The feds are providing pretty good resources, but not every case meets the criteria for district court. A lot of resources they have been giving go to waste because the USAOs still won't paper cases or DC Superior court Judges won't find probable cause at preliminary hearings or sign warrants. It's pretty crazy how much lower the bar is for probable cause in District court compared to superior court. The good news is the extra resources are trying.”
So it is entirely possible that the USAO has or will have good news to share with the public but as of now there’s little public evidence that the much-touted “surge” is making much of a difference relative to the enormous increases in arrests and prosecutions that happened before the DOJ stepped in. For instance, there have been very few announced “gang takedowns” or warrant sweeps over the last several months even though that is exactly the kind of thing Federal resources can do:
It has been ~10 months since the large “takedown” of the Kennedy Street Crew. This was an example of the long-term investigations in “drivers of gun violence” that USA Graves has said is a priority over more “reactive” prosecutions of arrests that MPD originates. “So you have two options: You can just wait and hope to catch one of them after they’ve committed an act of violence. Or you can figure out if they’re involved in other criminal conduct, such as illegal firearms possession or illegal firearms distribution, illegal narcotics distribution — and put out an investigation together around that conduct to try to get them off the streets.” Reviewing USAO press releases shows that almost all of the updates on “gang” or “crew” cases relate to arrests back in 2021, 2020 and 2018 now finally working up to a conviction. Obviously one can’t announce ongoing investigations but it’s concerning that there haven’t been more such takedowns given how many possible targets there are for DC investigators.
It has been ~6 months since the US Marshals and MPD’s much touted “Operation Trident” that arrested 48 people with outstanding warrants. While one administrator said “God I hope so” when asked if they thought these kinds of arrests would continue there haven’t been (to my knowledge) any similar announced operations in the meantime. Operation Trident was actually dwarfed by the 226 fugitives that the US Marshals helped apprehend in a similar operation in Baltimore so Federal agencies can make a big difference with warrant enforcement but so far have not announced comparable efforts in DC.
Washingtonians deserve more answers about how exactly this Federal “surge” of resources is helping make DC safer. One way is pressuring the US Marshals to share anything about their DC operations. The bar is incredibly low and so far the US Marshals have mostly evaded scrutiny for the many suspects arrested for new crimes who had had active (unenforced) bench warrants for months like this robbery suspect:
The largest accountability gap is related to the USAO’s efforts to prosecute “the individuals that we know to be driving violence.” Both the USAO and MPD agree that this is the priority population, as noted in MPD’s strategic plan: “Most of DC’s gun violence is driven by a relatively small group of people.” “MPD’s Violent Crime Suppression Division (VCSD) focuses on these individuals, identified as “Drivers of Gun Violence.” Given that this miniscule population is so critical, MPD and the USAO should be able provide some aggregated information on their efforts to bring them to justice:
How many individuals are currently identified as “Drivers of Gun Violence”?
What % of these identified individuals are currently facing charges?
What % of these individuals are currently detained?
What % of these individuals have an active warrant?
What % of these individuals are currently under investigation?
How many previously no-papered arrests of these individuals have been reviewed by prosecutors? How many are awaiting review?
Given that the USAO has significantly increased its prosecution rate for many offenses it is likely that some of these suspects’ older cases are actually viable; especially with a now-functioning crime lab. Identifying old cases to incapacitate these “drivers of gun violence” would require the USAO to admit some prior fault but would be good for public safety.
Mayor Bowser and the Council should ask for this information if for no other reason to gauge how these organizations are tracking this highest-risk population. The worry is that if MPD and the USAO can’t track these metrics for confidential reporting (though these aggregate numbers should be public) it would imply that they aren’t managing their resources in a prioritized way to address this population. Especially since the Bowser administration appears to be scaling back efforts at preventing violence (with a 54% proposed budget cut to Violence Intervention) it is even more important that the “stick” of increased enforcement be effective. The USAO especially hasn’t earned the benefit of the doubt and Washingtonians deserve proof that this focus on the “Drivers of Gun Violence” is more than just a talking point.
Always love this analysis. Despite not living near DC, the analysis is so good that this is one of my most looked-forward-to reads when it comes out.
Something I'm always curious about is those highest-risk 200-500 named suspects. Has anyone ever studied what happens to them over a given year? As in what % are arrested, jailed, charged? I have a feeling the death rate from violence and drug overdoses alone is probably shockingly high, but I guess I'm not sure. I guess what I'm ultimately wondering is how much this population is continually growing or shrinking even without legal action.