Trump claims crime is up in US cities. The truth is more complicated.

Trump has blamed the murder spike on Democratic mayors. But murders are up in Republican-run cities, too.

President Donald Trump and the Republican National Convention have characterized America as lawless and chaotic, with crime and violence rising in major cities. Trump has suggested this is a result of Democrats running the cities, arguing that “Democrat run cities are now rampant with crime.” Vice President Mike Pence made the pitch explicit at his RNC speech on Wednesday, saying, “The hard truth is you won’t be safe in Joe Biden’s America.”

Trump and Pence are right in one sense: One type of violent crime — murder — does seem to be up in large US cities this year compared to 2019. But violent crime is flat, compared to the previous year, and reported property crime is actually down, based on data from crime analyst Jeff Asher.

It’s not clear if this is a nationwide phenomenon, or if it’s isolated to urban centers, because we don’t have good data outside of the large cities.

But it’s not right to blame this exclusively on Democratic-run cities. Murders are also up in Jacksonville and Miami, both of which are overseen by Republican mayors and a Republican governor. And this is all happening under Trump’s presidency. The trend doesn’t appear to be partisan.

So what’s going on?

Some experts have cited the protests over the police killings of George Floyd and others — which could’ve had a range of effects, from officers pulling back from their duties to greater community distrust in police, leading to more unchecked violence. Others point to the bad economy. Another potential factor is a huge increase in gun purchases this year. Still others posit boredom and social displacement as a result of physical distancing leading people to cause more trouble.

Above all, though, experts caution it’s simply been a very unusual year with the Covid-19 pandemic. That makes it difficult to say what, exactly, is happening with crime rates. “The current year, 2020, is an extreme deviation from baseline — extreme,” Tracey Meares, founding director at the Justice Collaboratory at Yale Law School, told me.

That offers a bit of good news: It’s possible that the end of the pandemic will come and homicide rates will fall again, as they generally have for the past few decades in the US. But no one knows for sure if that will happen, or if we’re now seeing a shift in long-term trends.

Uncertainty about what’s going on isn’t exactly new in the field of criminal justice. Rates of crime and violence have plummeted over the past few decades in the US, yet there is no agreed-upon explanation for why. There are theories applying the best evidence, research, and data available, ranging from changes in policing to a drop in lead exposure to the rise of video games. But there’s no consensus.

That a decades-long phenomenon is still so hard to explain shows the need for humility before jumping to conclusions about the current trends.

“We don’t know nearly enough to know what’s going on at the given moment,” Jennifer Doleac, director of the Justice Tech Lab, told me. “The current moment is so unusual for so many different reasons that … it’s really hard to speculate about broad phenomena that are driving these trends when we’re not even sure if there’s a trend yet.”

All of that said, here’s what we do know.

Homicides are up this year in large US cities

There are several good sources, from criminologists, economists, and other data analysts, for what’s happened with crime and violence so far this year: an analysis by Jeff Asher; a Council on Criminal Justice report written by Richard Rosenfeld and Ernesto Lopez; and City Crime Stats, a website from the University of Pennsylvania set up by David Abrams, Priyanka Goonetilleke, Elizabeth Holmdahl, and Kathy Qian. They all focus on major cities, because we don’t have good data — and likely won’t until 2021’s federal reports — for other places.

Crime analyst Jeff Asher offers the most recent data, looking at violent crime and property crime trends in 25 US cities in 2020 so far compared to 2019. He found murders are up 26 percent, while violent crime is flat and reported property crimes are down. Overall, murder was up in 19 of 25 cities included in his analysis.

The Council on Criminal Justice report, published in July, looked at crimes in 27 US cities, ranging in size from New York to Cincinnati through June 2020. The authors looked for “structural breaks” in which reported crime increased or decreased more than would be expected, based on data from previous years.

They found structural breaks in homicide and aggravated assault increases starting in late May and June 2020, and structural breaks in robbery increases starting with the Covid-19 pandemic. But there weren’t statistically significant changes in gun assaults or domestic violence, though data was limited for the latter. And other kinds of crime, including larceny and drug offenses, trended down.

Here’s the graph for homicide increases, which were led by spikes in Chicago, Philadelphia, and Milwaukee, according to the report:

Council on Criminal Justice

That certainly suggests there were more homicides. But it’s hard to say if that’s a result of more shootings, as some news reports have suggested, given that numbers of reported gun assaults weren’t significantly different. It’s unclear what’s driving the increase in homicides if more shootings aren’t.

“It does look like violence is up in a number of cities,” Rosenfeld, one of the report authors, told me. “How widespread the increase is, it’s very difficult to know.”

City Crime Stats’ data complicates matters a bit, comparing the 2020 crime trends in 27 major cities to a five-year baseline. With this approach, the homicide increases don’t seem quite as dramatic in many cities, and other types of crime appear to be mostly down as well. Still, homicides do seem to be significantly up in many of the cities included in the City Crime Stats data set.

Here, for example, is Chicago, which shows this year’s rate (the red line) rising above the five-year baseline (the gray line and shading) at several points throughout the year:

City Crime Stats

There’s a lot of variation from city to city. Denver, Minneapolis, New York City, and Philadelphia are on the high end of homicides or seeing a flat-out increase. Austin, Baltimore, Boston, and Columbus are in line with historical trends or actually down.

Overall, though, Abrams said that his data suggests there was a significant increase in homicides from May to June: “We did find a statistically significant increase in homicides — about 21 percent — in aggregate in the cities we looked at in the month after versus before those protests,” he told me, cautioning that we can’t say with any confidence if the protests were the cause. “Same for shootings, but that’s from a smaller number of cities.”

In Chicago, as well as some other cities, the apparent increase in homicides began before the protests over the police killing of George Floyd. And in some cases, as in Chicago, the spike abruptly ended almost as quickly as it started, only to surge again weeks later, after the protests had died down. So it’s hard to blame only the protests for a spike — especially because we know that other factors likely played a role, such as the start of summer, when crime tends to go up, and the end of stay-at-home orders.

City-by-city variation isn’t unique to 2020. It’s expected, even when talking about national crime waves or declines, to see some places go up and others go down for different kinds of crime. The US is a big country, and a range of local factors can affect different kinds of crime.

Still, there’s enough in the three data sets to draw some conclusions: At least in major US cities, homicides are up overall this summer — in some cases, significantly higher. But other kinds of crime, including violent crime overall, aren’t up and may actually have decreased so far this year. There was also a brief spike in burglaries in major cities starting in late May — an increase that was so brief and contained to specific cities that experts told me it was likely due to riots and looting surrounding some Black Lives Matter protests.

As Asher noted on Twitter, a disconnect between murders and other crimes would be odd: “Violent crime and murder almost always move in the same direction and they are never this far apart nationally.”

One way to reconcile this may be the nature of crime reporting. All of this data is based on reports to governments, typically local police departments. But with people stuck at home, and no government agency operating normally this year, perhaps these reports are just less likely to happen or get picked up this year, especially lower-level crimes involving drugs or stolen property.

At the same time, it’s far harder for a homicide to go completely unreported — it’s difficult to ignore a dead person. This is why, for much of US history, the homicide rate has been used as a proxy for violent crime overall: The nature of homicide made it a more reliable metric than others for crime.

In other words, it’s possible that other kinds of crime are up this year, but they’re simply going unreported. At any rate, homicide does seem to be up overall, at least in major US cities.

One note on domestic violence: Some activists and experts worried it would increase this year as people were forced to stay home more often. The Council on Criminal Justice report and City Crime Stats’ analysis suggest that’s not the case, showing no significant change or a drop in some places. But there’s reason for skepticism: Both sources are pulling data from a limited number of cities. And reporting limitations may especially apply to domestic violence, since this year victims are potentially more likely to be trapped with their abusers and unable to make a phone call for help.

There are plenty of caveats to all this data. It only represents the trends in large US cities, which means it might not be representative of the country as a whole. And it only covers 2020 through June, July, or August, depending on the report, so there are four to six months for the captured trends to change.

But the trend in some places, particularly with homicides, is alarming.

We know less about why there might be a spike, but there are some theories

So why did homicides increase in some cities?

When I posed this question to experts, they again cautioned that no one can say with certainty what’s going on. That said, they offered some possible explanations, based on the limited information we have so far:

1) The pandemic has really messed things up: Looming over absolutely every discussion about 2020 is the Covid-19 pandemic. That’s no different for discussions about crime and violence. This year is very unusual, with many forced to stay at home and living in fear of a new, deadly virus. That could lead to all sorts of unpredictable behaviors that experts don’t understand yet, and that might take us years to explain.

2) Depolicing led to more violence: In response to the 2014 and 2015 waves of Black Lives Matter protests against police brutality, officers in some cities pulled back, either out of fear that any act of aggressive policing could get them in trouble or in a counter-protest against Black Lives Matter. While protesters have challenged the crime-fighting effectiveness of police, there is a sizable body of evidence that more, and certain kinds of, policing do lead to less crime. Given that, some experts said that depolicing in response to protests could have led to more violence — what some in years past called the “Ferguson effect,” after the 2014 protests in Ferguson, Missouri, over the police shooting of Michael Brown, and also seen in Baltimore after the 2015 killing of Freddie Gray.

3) Lack of trust in police led to more violence: In response to the “Ferguson effect” in 2015, some experts offered a different view of what was happening: Maybe people had lost trust in the police and, as a result, they relied more on street justice and other illegal activities to resolve interpersonal disputes — an interpretation of “legal cynicism,” explained well in Jill Leovy’s Ghettoside and supported by some empirical research. Perhaps Floyd’s murder and the ensuing protests led to a similar phenomenon in some cities this year.

4) More guns led to more gun violence: There’s been a massive surge in gun buying this year, seemingly in response to concerns about personal safety during a pandemic. And as the research has shown time and time again, more guns mean more gun violence. A recent, preliminary study from researchers at UC Davis already concluded that gun purchases led to more gun violence than there would be otherwise through May this year. That could have further exacerbated homicide increases.

5) Overwhelmed hospitals led to more deaths: One way to explain a flat or dropping violent crime rate as homicides rise is that the violent crime was deadlier than usual. With health care systems across the US at times close to capacity or at capacity due to Covid-19, maybe hospitals and their staff had less ability to treat violent crime victims — increasing the chances they died this year. That could translate to more deaths, and homicides, even if violent crime remained flat or declined.

6) Idle hands led to more violence: Throughout the pandemic, a lot of people have been bored — with forms of entertainment, from restaurants to movie theaters, closed down. Schools are shut down too, and millions are newly unemployed. Other support programs that can prevent violence were shuttered due to the lockdowns. All of that could have led to conflict, and possibly more crime and violence. But, experts cautioned, this is speculative, with little evidence so far to support it.

7) A bad economy led to more violence: With the economy tanking this year, some people may have been pushed to desperate acts to make ends meet. Disruptions in the drug market, as product and customers dried up in a bad economy, may have led to more violent competition over what’s left. The bad economy also left local and state governments with less funding for social supports that can keep people out of trouble. All of that, and more, could have contributed to more crime and violence, but this, too, is still very speculative.

Another possibility: None of these explanations is right. With limited data in strange times, it wouldn’t be surprising if it turns out we have no idea what’s going on right now. “We can bet on it being unpredictable,” Doleac said.

Again, there’s still no consensus about what’s caused crime to decline since the 1990s. In that context, it’s no surprise there’s nowhere near a consensus as to why a homicide spike that may not even be a national or long-term phenomenon has occurred so far this summer.

The trends could change after a strange 2020

It’s possible that, before we understand why it’s happening, the year’s alarming homicide trends could recede. It’s happened before: In 2005 and 2006, the homicide rate briefly increased, only to start declining again before hitting record lows in 2014. In 2015 and 2016, the rates also spiked again only to start to dip after. In both instances, these years were effectively blips and the overall crime decline America has seen for the past three decades continued.

Maybe after this very weird year ends, crime and violence trends will, similarly, go back to the previous normal.

But that’s not a guarantee — and it’s not something, experts said, that we should rely on. “We don’t really understand why crime and violence went down,” John Roman, a criminal justice expert at NORC at the University of Chicago, told me. “Being able to say we should expect this unexplained phenomenon to continue strikes me as sort of irrational.”

Even if we can’t explain what may be causing a homicide spike in some cities, there are certain strategies that might help fight crime in the short term — such as deploying police in crime hot spots (though that would have to be done carefully and with reforms, given the current political climate around policing), a “focused deterrence” program that targets the few people in a community engaging in violence with a mix of support and sanctions, and using civilian “interrupters” to personally intervene in cases in which violence seems likely to break out.

Notably, a lot of this work is done at the local and state level, where the vast majority of police departments are based. The federal government can incentivize certain practices, like Biden has proposed doing, but it ultimately falls on cities, counties, and states to carry out new or revised approaches.

Many of the evidence-based approaches rely on in-person contact, which requires ending the pandemic. “The police, public health, and community approaches to violence reduction require that people meet face-to-face; they cannot be replaced by Zoom,” Rosenfeld and Lopez wrote in the Council on Criminal Justice report. “An underappreciated consequence of the pandemic is how social-distancing requirements have affected outreach to high-risk individuals.”

So priority number one should be to end the pandemic — ending its potential ripple effects on crime and enabling evidence-based approaches that can help reduce crime. But to do that, the US public and governments will need to truly embrace strategies that have worked for countries like South Korea and Germany against Covid-19: physical distancing, masking, and testing, tracing, and isolating the sick. In this sense, Trump’s failures to address Covid may be leading to more violence.

“Seeing what’s happening with these [crime] numbers can point us to or at least get us thinking about what potential policy levers we could employ that would be helpful,” Doleac said. “Otherwise, our attention is probably better focused on making sure we’re all wearing masks.”

Beyond the pandemic, police are going to have more trouble fighting crime — including any current or future spikes — if large segments of the community don’t trust them. That’s where police reform comes into play. It’s a complicated topic, separate from a possible spike in violence this year. But, in short, experts say police should, at a minimum, show the communities they serve that they understand the concerns, acknowledge mistakes, and will change how officers are deployed and targeted.

Otherwise, there’s a good chance that protests against police will flare up, just as they did from 2014 to 2016 and have again this summer. If protests lead to more violence — whether by leading to depolicing, or sowing and exposing distrust in law enforcement — that’s going to create public safety problems.

To put it another way: There’s a lot we don’t know about crime, why it happens, and how to stop it. But it’s going to be much easier to wrap our heads around these issues once things get closer to how they should be — and that means seriously addressing the pandemic and protests against police brutality.

Unfortunately, the US is going in the opposite direction, with the recent resurgence of the coronavirus and Trump exacerbating police-community tensions with his rhetoric and push to deploy unsolicited federal agents in US cities.

“How optimistic should we be for the rest of the summer?” Roman said. “I think the answer is not terribly optimistic, because none of these factors seem to be abating with the return of Covid.”

Support Vox’s explanatory journalism

Every day at Vox, we aim to answer your most important questions and provide you, and our audience around the world, with information that has the power to save lives. Our mission has never been more vital than it is in this moment: to empower you through understanding. Vox’s work is reaching more people than ever, but our distinctive brand of explanatory journalism takes resources — particularly during a pandemic and an economic downturn. Your financial contribution will not constitute a donation, but it will enable our staff to continue to offer free articles, videos, and podcasts at the quality and volume that this moment requires. Please consider making a contribution to Vox today.