Why America's Great Crime Decline Is Over

This story was originally published by The Atlantic. Subscribe to the magazine’s newsletters.

Americans are experiencing a crime wave unlike anything we’ve seen this century. After decades of decline, shootings have surged in the past few years. In 2020, gun deaths reached their highest point in U.S. history in the midst of a pandemic. In 2021, although researchers can’t yet say anything definite about overall crime, shooting incidents appear to be on the rise in many places. We have also already witnessed several mass shootings, including the murder of spa and massage workers in the Atlanta area and a grocery-store massacre in Boulder, Colorado. Americans can no longer say, as we could 10 years ago, that we are living in the safest time in our nation’s history.

Why crime rises and falls is a devilishly complicated question. Few people have thought more deeply about it than Patrick Sharkey, a sociologist at Princeton University. While others reach for easy solutions and simplistic slogans, Sharkey embraces complexity and uncertainty. In his 2018 book, Uneasy Peace, Sharkey argued that intensive and often aggressive policing and incarceration policies probably helped reduce crime in the past few decades, to the great benefit of low-income neighborhoods. But rather than glorify these policies, he argued that often they have involved brutal policing strategies that could provoke a backlash among the public—hence the “uneasy” nature of the peace.

[Alec McGillis: How to stop a police pullback]

This thesis has proved doubly prescient in the past year. Sharkey anticipated both the summer of anti-police protests and the possibility that souring police-civilian relations would contribute to an increase in violent crime.

This week I spoke with Sharkey about his thoughts on the 2020 crime surge. Our conversation has been edited for length and clarity; statistical context from a follow-up email is in italics.

Derek Thompson: What happened last year?

Patrick Sharkey: It was a huge surge of violence, and the most violent year of the century. We went through a long period where violence was steadily falling. There was a sharp decline in the 1990s and a more gradual decline since then. But right now we are in a period of rising violence. Since 2014, there has been a gradual increase. And then last year was a really terrible year across the whole country.

Thompson: The subtitle of your book Uneasy Peace is The Great Crime Decline, the Renewal of City Life, and the Next War on Violence. Is it safe to say that the “great crime decline” has come to an end?

Sharkey: I would say it is very clearly paused. What remains to be seen is just how anomalous last year was. There’s a possibility that this was just a year when social life was completely destabilized in so many ways, and that resulted in a huge surge of violence that was temporary. That’s the hope.

Thompson: Where is crime rising today?

Sharkey: This is the analysis I’m doing right now. It’s always been true that violence is concentrated in a small number of communities. The current increase in crime is not evenly distributed, either. Most of the increase in violence is highly concentrated in neighborhoods that are segregated with high poverty. Many of these neighborhoods have experienced disinvestment for generations, for decades, and it has made them more vulnerable to violence. Their public spaces have not been maintained. Their schools are underfunded. Their parks are not maintained. There aren’t functioning community centers or after-school programs for children.

In research shared exclusively with The Atlantic after we spoke, Sharkey calculated that Chicago had 267 more fatal shootings in 2020 than the previous year. This was by far the largest numerical increase in the country and more than double that in any other city. New York City, Philadelphia, and Atlanta saw fatal shootings increase in 2020 by more than 30 percent. But in several other major cities—including Miami; Las Vegas; San Jose, California; and Honolulu—fatal shootings declined, according to his data.

Thompson: Do we know if certain groups are suffering disproportionately from this crime increase? There have been several news reports of increased harassment and crime against Asian Americans, even before the Atlanta-area massacre. Is there anything your data can tell us about whether violence is increasing for some ethnic groups more than others?

[Read: What caused the great crime decline in the U.S.?]

Sharkey: My work looks most closely at where crime is happening, not at individual victims. But there are some things we think we know. Intimate-partner violence increased in 2020. So did hate crimes against Asians. But the overall demographics of victims is incredibly consistent over time. It’s young people of color, particularly young men of color. I don’t see anything yet to indicate that’s changed dramatically.

Thompson: A lot of books and stories have been written about the great crime decline in the U.S. over the past 30 years. Now that this decline seems to be on pause, I wonder how you think some of these theories hold up, or don’t.

Sharkey: So much research and journalism on violence gravitates toward single-cause explanations. And that’s just misguided. We know that violence is incredibly complex, but we consistently seek out counterintuitive explanations. Take lead exposure, for example, which was a popular single-cause explanation for the decline in crime. Look, lead matters for crime. But you can’t show graphs of how violence rises and falls within the 2010s and say this is purely explained by a change in lead exposure from 20 years ago. That’s simplistic and incorrect.

Thompson: I want you to help me understand the relationship between poverty and crime, as you see it. There seems to be an association between high-poverty areas and high crime. But in the past 13 years, we’ve had two significant nationwide recessions. The first downturn coincided with falling crime. The second downturn coincided with the largest increase in crime in decades. What should that teach us?

Sharkey: I’ve argued against the idea that it’s about individual economic conditions. It’s very clear that economic recessions typically do not translate into more crime.

My argument is that in areas where communities go through periods of disinvestment and where institutions break down, people feel like they’re on their own. This creates conditions where violence becomes more likely. As a place becomes more violent, people change their behavior. They become more likely to interpret uncertainty in an aggressive way, more likely to carry a weapon, more likely to act quickly or first if they feel threatened. This is how the presence of violence creates more violence. This cascading effect, where violence begets violence, has been reinforced in the past year.

Last year, everyday patterns of life broke down. Schools shut down. Young people were on their own. There was a widespread sense of a crisis and a surge in gun ownership. People stopped making their way to institutions that they know and where they spend their time. That type of destabilization is what creates the conditions for violence to emerge.

Thompson: To what extent should we blame the lockdowns, specifically, for separating people from these institutions? A lot of COVID-19 restrictions effectively shut down libraries, schools, and other places. I wonder if you think this idle time was a multiplier for violence.

[Patrick Sharkey: To avoid integration, Americans built barricades in urban space]

Sharkey: It’s not just idle time but disconnection. That might be the better way to talk about it. People lost connections to institutions of community life, which include school, summer jobs programs, pools, and libraries. Those are the institutions that create connections between members of communities, especially for young people. When individuals are not connected to those institutions, then they’re out in public spaces, often without adults present. And while that dynamic doesn’t always lead to a rise in violence, it can.

Thompson: We have to talk about the effect of the 2020 summer protests on the rise in violence. I can imagine agreeing with two very different arguments. There’s the left’s argument that it would be reckless to exclusively attribute the increase in violent crime to the Black Lives Matter and “Defund the police” protests, which were understandable responses to very real problems. Clearly, last year was destabilizing in so many other ways. But then there’s the more conservative, I suppose you could call it, argument that the “Defund the police” movement was substantively wrong and possibly counterproductive, because more policing, not less policing, has been commonly associated with crime reduction. What’s more, the protests may have directly contributed to a pullback in police presence in certain cities. And sudden declines in police-civilian interactions have been associated with large increases in local crime. What’s the right way to sort through these interpretations?

Sharkey: Violence surged last year after the protests. There’s no doubt about that. The important part is to interpret it.

You’re right that, in the aftermath of high-profile [police scandals], the police often pull back. They change the types of incidents they get involved in. This can be political, or it can be individual choices. But there’s a second part. Residents and community members also often step back in the aftermath of high-profile incidents when the legitimacy of police departments is questioned. The public becomes less willing to reach out to police for help and less willing to cooperate or provide information.

When a social order depends on the police dominating public spaces, and that form of social order is questioned and starts to break down, it can lead to a surge in violence. It doesn’t mean that protests cause violence. It means that when you depend on the police to dominate public spaces and they suddenly step back from that role, violence can increase.

Thompson: I want to make sure I understand your position here. One argument says that the crime surge followed the protests, so we can safely assume that the protests at least partly contributed to more crime. You’re saying if the crime surge is downstream from the protests, the protests are downstream from something else—an unstable social order. We have to go upstream to fix the public’s relationship with policing.

Sharkey: That’s right. In my book I argued that the drop in violence had all these benefits, but it was unsustainable. And it was unsustainable because we were reliant on a model of responding to violence and urban inequality through brute force and punishment and dependence on prisons and the police to respond to every challenge that arises when poverty is concentrated.

As long as that model is still in place, you can produce lower levels of violence. But you will also produce staggering levels of harm. That harm became very visible in the last five years and, in particular, after [the killing of] George Floyd.

Thompson: Some people say that the simplest way to reduce crime will be to re-surge police presence in violent areas. I’ve called for “unbundling” police services and paying a broader group of local government and state employees to work with the homeless or handle domestic disputes. How do you think we should begin to reverse the rise in crime?

Sharkey: These surges in violence came from an old model of policing. In this old model, police respond to violence with brute force. And this can reduce violence. But it comes with these costs that don’t in the long run create safe, strong, or stable communities. We shouldn’t reach back for the old model and forget that the old model actually came with all this resentment and harm. This is our wake-up call. We need a new model.

Police still need to play a role in that model, because the U.S. has the problem of guns, and there aren’t many organizations other than police departments that can deal with the problem of gun violence. Police still have an essential role to play. But we need to invest in other local organizations and residents. We need to give these groups the commitment that we’ve previously reserved for the corrections system and law-enforcement agencies.