The year 2020 saw the largest recorded increase in homicides in United States history — an increase likely propelled by a complex mix of factors, from more guns to stresses of the pandemic to fewer police officers on the streets to a crisis in relations between police and citizens.

But one persistent theory is that a change in policing last summer primarily drove increased gun violence. This is an especially popular explanation among law enforcement figures. Former Baltimore Police Department Deputy Commissioner Jason Johnson recently argued that the real driver of last year’s murder rise was a severe decline in police activity, especially after protests erupted last summer in the wake of George Floyd’s murder. 

St. Louis Police Commissioner John Hayden suggested that the police resources devoted to protests prevented officers from engaging in neighborhood policing. Former NYPD Commissioner Ray Kelly said police were “stretched to the limit” by the protests and coronavirus restrictions. Summarizing widespread reductions in stops and arrests, Johnson wrote that “when the Thin Blue Line retreats, violence charges in.”

But data from numerous large American cities complicates that narrative, suggesting that the change in policing alone is not sufficient to explain last year’s large increase in murder and that a growing number of firearms on the streets likely played a significant role.

It’s true that police activity, as measured by stops and arrests, declined significantly in 2020. Still, despite that drop, and weeks before Floyd’s murder and the ensuing protests, police began finding firearms more often than in previous years.

This pattern does not support the idea that overwhelmed police forces weren’t able to take guns off the streets, leading to a surge in violence. Instead, the spike in firearms as a percentage of stops and arrests provides evidence that there were simply more guns on the streets throughout 2020 than in the past, which may have intensified other sources of violence and contributed to the historic rise in murders.

While there is no standardized, national open data on stops, information on police activity in 10 cities that we compiled points toward the same pattern.

First, stops and arrests fell rapidly in each city in March and April 2020, driven by pandemic restrictions on police contact or due to fewer people being outside (and thus available to be stopped by police). 

If less policing alone led to increased violence, we would have expected to see an uptick in March and April after this clear change. But there was no observable increase in gun violence in these cities at that time.

Police activity dropped again after Derek Chauvin murdered George Floyd in late May 2020, this time with an accompanying surge in shootings in many cities. Cities generally saw stops and arrests increase over the last few months of 2020 — though still below pre-pandemic levels — with the elevated level of violence remaining. 

While the volume of stops and arrests fell dramatically in March and April in all 10 cities, police in every city were more likely to find a firearm when they made stops and arrests. In Chicago, for example, police stops decreased nearly 70 percent between January and May 2020, but officers actually found 83 percent more firearms in May than in January.

Jens Ludwig, director of the University of Chicago Crime Lab, analyzed stops in Chicago and concluded that “unless the police have become dramatically better at figuring out who is illegally carrying a gun (and so have become better at figuring out who to stop), the implication is that lots more people are carrying guns illegally in Chicago.”

The same pattern was seen across numerous cities with available data. There were 34 percent fewer arrest charges in Los Angeles in April and May 2020 compared to April and May 2019, but charges for weapons possession were up. The problem was not confined just to big cities, either. In Tucson, Arizona, for example, there were 39 percent fewer arrests in April and May 2020 compared to a year earlier but 29 percent more arrests for weapons or firearms possession. 

The share of stops or arrests that resulted in a firearm being found increased in every city. In Washington, DC, the share of all arrests that were weapons violations went from 5 percent in January to March 2020, to 7 percent in April and 9 percent in May. The share of arrests for weapons possession went from 1 percent between January and March 2020 in Charleston, South Carolina, to 4 percent between April and December. 

Almost every city followed the same pattern: a dramatic jump in the share of arrests or stops with a firearm in April and May, a decline in June, and a return to the earlier elevated levels for the remainder of the year.

The legitimacy crisis in law enforcement

The implication of this trend is that — assuming police did not suddenly become substantially better at identifying who has an illegal gun — firearm carrying increased at the beginning of the pandemic, well before the protests, and persisted at that level for the remainder of the year. 

It is possible that in the midst of the pandemic, police started engaging in better-targeted stops that were more likely to yield arrests. But finding other kinds of contraband, like drugs, did not become more frequent, only guns.

Data on investigatory stops — defined as stops “based upon reasonable suspicion that the person has committed, is committing, or is about to commit a crime” — in Chicago is instructive and suggests more firearms were found because more were being carried, rather than a change in policing strategy.

The share of searches in investigative stops that found drugs just before Covid-19 lockdowns was virtually unchanged after Covid-19, going from 20.9 percent between October 2019 and March 2020 to 20.7 percent between April and September 2020. The demographics of searches did not change much, either, with Black people making up 74.3 percent of people searched in stops from October 2019 to March 2020 and 76.1 percent from April through December. But CPD officers found firearms in 11.5 percent of searches from April to September, compared to 3.7 percent of searches in the six months prior.

Since all cities with data had an increase in the share of stops or arrests with a gun at around the same time, no one change in departmental or prosecutorial policy can explain why.

Investigative stops and arrests show an increase in firearm carrying beginning in March or April, shortly after background checks surged to unprecedented levels nationally. More firearms could have contributed to the historic rise in murders in 2020 by turning less dangerous crimes into potentially lethal encounters. 

Police finding more firearms in stops and arrests does not fit with the idea that a decrease in proactive police activity targeting firearms was the major driver for 2020’s historic murder totals, though it certainly cannot be ruled out as a contributing factor. 

Johnson put the blame on progressive prosecutors, writing that “making arrests for drug and weapons crimes that will go unprosecuted exposes officers to the risk of disciplinary action, lawsuits and criminal prosecution. To mitigate that risk, police take a more passive approach.” But firearm arrests increased 42 percent in Philadelphia — home of progressive prosecutor Larry Krasner — between April and December 2020, compared to the same time frame in 2019.

The data all points to substantially more complex causes behind the rise in murder than the simple narrative of a change in policing as the sole or even main driver. It is plausible, though, that the summer’s drops in stops and arrests, protests against police violence, and increases in gun violence are all symptoms of the same disease: what criminologists David Pyrooz, Justin Nix, and Scott Wolfe recently called a “legitimacy crisis in the criminal justice system,” the result of intensifying distrust in “the law and its gatekeepers” as a result of injustice.

Writing in the Denver Post, they said that a “legitimacy crisis is consequential for three reasons. The first is depolicing, where officers pull back from proactive policing in response to public criticism. Second, depleted trust in the law means citizens will think twice about calling the police to report crimes or suspicious behaviors. Lastly, delegitimacy of the law emboldens criminal offending populations, as the moral obligation to follow the law is weakened.”

The trend toward more firearms sales and more guns on the street seems to have continued into 2021. Background checks accelerated even beyond last year’s peak in the first three months of this year. And the latest data from these cities’ stops shows that police are finding as many guns as they did in the second half of 2020.

Early figures from many cities show murders have increased from last year’s baseline as well. If the greater availability of firearms contributed to last year’s violence, the latest arrest data suggests it may contribute even more deaths to 2021’s murder total.

Rob Arthur is an independent journalist and data scientist based in Chicago. He’s on Twitter at @No_Little_Plans. Jeff Asher is a crime analyst based in New Orleans and co-founder of AH Datalytics. You can find him on Twitter at @Crimealytics.