:format(webp)/cdn.vox-cdn.com/uploads/chorus_image/image/56839543/52349842.0.jpg)
It’s now official: After several decades of drops, America’s murder rate increased for the second year in a row in 2016.
The FBI put out its official crime numbers for last year over the weekend. They tell a very similar story to that of 2015. The murder rate increased by 8 percent over 2015, while the violent crime rate, including rape and robbery, increased by more than 3 percent. The property crime rate, however, decreased — by about 2 percent.
As criminologist John Pfaff pointed out, most of the increase in the murder rate in particular — which is widely considered the most accurate proxy for crime — was linked to cities with a population of 250,000 or more. Chicago alone contributed to about 22 percent of the increase in murders.
One point of caution: Crime is still below what it was several years ago. Even at 5.3 per 100,000 people, the murder rate, for example, is still below what it was in 2008 and the years before that, and it’s nearly half of what it was during its peak in 1980.
The increases also don’t mean that crime and murder have gone up across the board, even in urban areas. New York City, for example, has seen crime drop or hold steady for years.
Criminologists don’t have a definitive answer yet as to why the overall national figures increased — and they caution that much of this still needs serious study to figure out.
But after the second year in a row of national increases, the report raises a red flag about what is happening in the US. Could crime be back on the rise? Or are the 2015 and 2016 numbers, like similar increases in 2005 and 2006, just another temporary rise before the crime and murder rates continue their long-term decline? And what could be behind the recent increases?
One theory for the increase: the “Ferguson effect”
Criminologists caution that it’s probably still too early to know why violent crime and murder have been going up in the past couple of years, but they have some early ideas.
One prominent theory is what’s known as the “Ferguson effect”: the impact of Black Lives Matter protests against police shootings since Michael Brown’s 2014 death in Ferguson, Missouri.
There are essentially two hypothesized versions of the Ferguson effect: One is that Black Lives Matter protests have scared law enforcement officers from doing the proactive policing necessary to prevent crime, while at the same time criminals have been emboldened because they now know police are backing off from aggressive tactics. The other theory is that these protests have reinforced communities’ distrust of law enforcement, making it harder to solve and prevent crimes. It’s also possible both of these versions are playing a mixed role.
Conservative columnist Heather Mac Donald has been a major proponent of the first version of this theory. She’s pointed to data showing that after protests and riots over the death of Freddie Gray in Baltimore in 2015, murder rates began to climb. And there was some evidence in her favor in the data for last year, with violent crime and murder rates in Chicago rising after protests over the police shooting of Laquan McDonald. (But we’ll need a more rigorous analysis before this connection is validated.)
Richard Rosenfeld, a criminologist at the University of Missouri St. Louis who studied the 2015 increase in murders, has cautiously argued that the other version of the Ferguson effect is more plausible. He previously told me, “There is good sociological and historical evidence that if people lose confidence in the police to protect them, or if they lack trust in the police because they believe the police are harassing them or behaving unfairly, they do tend to take matters into their own hands. So one tends to see preemptive killings and retaliatory shootings go up.”
The distinction matters for policy. If Mac Donald’s version of the Ferguson effect is right, it vindicates proactive, aggressive policing tactics that have characterized the US’s “tough on crime” policy era. If the other version is right, it calls for less aggressive tactics and more community engagement.
Still, there’s reason for caution: No rigorous study so far has definitively backed up either version of the Ferguson effect. And its impact may not last long: For example, Baltimore’s murder rate declined in 2016 after a massive increase in 2015.
The increase in crime rates may be temporary
Another possibility, echoed by every criminologist I’ve talked to about this, is that the 2015 and 2016 changes are short-term fluctuations that won’t become a long-term trend.
We’ve seen short-term fluctuations before. The murder rate went up steadily from 5.5 per 100,000 people in 2004 to 5.6 in 2005 and 5.8 in 2006 — before dropping all the way down to the historical low of 4.4 in 2014.
The same thing could have happened in 2015 and 2016. For whatever reason (maybe even some version of the Ferguson effect), the two years were just particularly bad for murders and crime. But maybe it was just a blip, and the long-term trend of the crime decline will continue.
Preliminary figures for 2017 suggest this could be the case this time around as well. A recent report by the Brennan Center for Justice found that the murder rate decreased by 2.5 percent and the violent crime rate by 0.6 percent in the 30 largest US cities so far throughout 2017. Although the numbers are preliminary and not official, Brennan’s reports have been fairly accurate in providing an early look at crime trends over the past several years. (They predicted the murder rate for 2016 nearly four months ago.)
It’s also possible that the murder rate alone gives a distorted view of violent crime. Since the murder rate is generally very low, it’s prone to large statistical fluctuations. As one example, New Orleans–based crime analyst Jeff Asher previously told me he expected the 2016 mass shooting at a gay nightclub in Orlando, Florida, in which 50 died, to lead to a massive increase in the murder rate in the city, even though it was just one particularly bad event.
The increase may be caused by something we don’t know about
Consider this: Criminologists are still debating what caused the roughly 50 percent decrease in violent crime and murders since the 1980s and ’90s. There are many theories, from changes in policing tactics to more incarceration to reductions in lead exposure. But no one knows for certain.
Similarly, experts are beginning to develop all sorts of theories for why there may be an ongoing rise in murder and violent crime rates. Asher, for one, told me it’s possible police are beginning to pull back on tactics and strategies that worked to reduce crime before, perhaps as a result of dwindling resources. University of Chicago criminal justice researcher John Roman argued that the increase in murders may be driven by shooters using higher-caliber weapons — a claim backed by recent reports — while a rise in shootings in Chicago could be driven by a contagion of shootings leading to retaliatory shootings.
But again, all of this needs more study as researchers work through hyper-local trends to tease out a plausible national story. In fact, it’s possible the ultimate cause is a factor we haven’t even considered yet. Or there may not be a single national driver for the increases at all; instead, the increases may be driven by disparate local issues that vary from city to city.
One point that criminologists repeatedly emphasize in this field is that the data is very messy and noisy — sometimes even contradictory. For example, while the FBI last year found that violent crime went up in 2015, the National Crime Victimization Survey for the same year indicated that violent crime actually decreased. This is one of the major reasons that criminal justice experts caution against sweeping conclusions from one or two years of crime data: When the data can be all over the place, it’s better to wait for a few more years of statistics to make sure the figures really hold up in the long term.
Statistical fluctuations and “we don’t really know yet” aren’t the most satisfying answers. But given that they’re seriously plausible, everyone should probably wait a few years before trying to build a definitive theory around one or two years of shoddy crime statistics.
