More people were killed last year than in 2014, and no one’s sure why

The number of homicides in the country's 50 largest cities rose nearly 17 percent last year, the greatest increase in lethal violence in a quarter century.

A Wonkblog analysis of preliminary crime data found that about 770 more people were killed in major cities last year than the year before, the worst annual change since 1990.

The killings increased as some law enforcement officials and conservative commentators were warning that violent crime was on the rise amid a climate of hostility toward police. They said protests and intense scrutiny of officers who used lethal force had caused officers to become disengaged from their jobs, making streets more dangerous. Some have called it the "Ferguson effect," after the St. Louis suburb in which Michael Brown Jr. was shot and killed by a police officer in 2014.

A closer look at the figures, however, suggests no single explanation for the increases and reveals no clear pattern among those cities that experienced the most horrific violence.

[We've had a massive decline in gun violence in the United States. Here’s why.]

Several cities that recorded the largest increases in homicides -- Nashville and Washington, D.C., for instance -- had no widely publicized, racially charged killings by police. Many other big cities recorded modest increases or even declines in the number of homicides, with no deviation from the pattern of recent years.

Also, undermining the theory that police have become generally disengaged, a preliminary FBI report released last week showed that the overall number of violent offenses increased just 1.7 percent nationally during the first half of the year while the number of property crimes declined 4.2 percent.

(The FBI's official count for the full year will be published in the fall. Because a crime's classification may change as authorities gather more information, the official figures on homicide might differ from Wonkblog's current tally.)

[It's official: There never was a 'war on cops']

Public safety has been improving for two decades, and lethal violence in large cities is still rare by historical standards. Twice as many people were killed in those 50 cities in 1991 as in 2015. "You certainly wouldn't want to say the sky is falling," said Darrel Stephens, executive director of the Major Cities Chiefs Association.

Nonetheless, last year's interruption in the decline in homicides has experts concerned. They say it's too early to know what caused the change, or whether it will endure. It's not clear if there is a Ferguson effect, or if the homicides are a result of the heroin epidemic, reduced police department budgets, a decline in the number of convicts behind bars or other factors entirely.

"There's no national pattern," said Franklin Zimring, a criminologist at the University of California at Berkeley.

The Ferguson effect

When law enforcement officials from around the country met in Washington in October to discuss the increase in murders, many suggested that deteriorating relations between police and civilians could be the cause.

With civilians recording their every move with cellphone cameras, police may be hesitating to engage with criminals, several officials said. Perhaps criticism from civic leaders has damaged officers' morale, or maybe monitoring protests and marches draws crucial manpower away from solving cases.

"Has policing changed in the YouTube era?" asked FBI Director James B. Comey. "Cities with nothing in common are seeing – in the same degree and in the same time – dramatic increases in violence, especially homicides."

[How Flint, Ferguson and Baltimore are all connected]

The data does not provide clear evidence for this theory on a national level, but in some cities, intensified mistrust of the police appears related to spates of violence last year.

In Baltimore, where riots followed the death of Freddie Gray in police custody in April, the number of homicides increased 59 percent last year, from 217 to 344. That is more homicides than have been committed in the city in any year since 1993, when the city's population was 715,000, according to the census. Only 623,000 people live there today.

Several officers have been charged in connection with Gray's death. The case against one of them, William G. Porter, ended with a hung jury and a mistrial last month. Prosecutors said Porter was culpable in Gray's death because he failed to properly buckle him into the back of a police van, in which Gray suffered a fatal injury to his neck, and because Porter should have sought medical help before it was too late.

Peter Moskos, a former Baltimore police officer who teaches at the John Jay College of Criminal Justice in New York, disagrees. "These cops are on trial, arguably, for doing nothing wrong," he said, adding that such cases could discourage police from making arrests.

"That idea, that 'if I do my job, I could get in trouble' -- that has a chilling effect," Moskos said.

In Cleveland, anguish and protests followed the shooting death of 12-year-old Tamir Rice at the hands of a police officer in November 2014. The number of homicides nearly doubled in that city the next year, from 63 in 2014 to 120. Last month, a grand jury declined to charge the officer who shot Rice.

No correlation

At the same time, some of the most drastic increases in homicides occurred in cities where there have not been major controversies over law enforcement practices.

The number of homicides increased 83 percent in Nashville last year, to 75. The figure increased 62 percent in Oklahoma City, to 73. There were 162 homicides in Washington, D.C., last year, an increase of 54 percent.

[Washington Post database: Tracking homicides in the D.C. region]

These are all substantial changes, even though relations between police and civilians in these cities don't appear to have worsened last year.

"There doesn’t seem to be, to me at least, a correlation between where there were protests against the police, and where murder seems to be increasing," said Inimai Chettiar, director of the justice program at New York University's Brennan Center for Justice.

While protesters have marched against police violence in nearly every U.S. city, the acrimony has been greater in some cities than in others. Indeed, in Nashville, officers offered protesters hot chocolate and coffee.

"I personally don't really believe that police officers across America have pulled back from doing their job," said Stephens of the Major Cities Chiefs Association, who also served as police chief in Charlotte for nine years.

In other cities, homicides didn't escalate significantly last year, despite intensifying conflicts with civilian leaders.

UC Berkeley's Zimring points to New York, where relations between police and the city as a whole have largely collapsed.

Mayor Bill de Blasio won election on a promise to overhaul the police department after a federal judge ruled the widely controversial practice known as "stop and frisk" unconstitutional. Then, a grand jury didn't indict an officer for the death of Eric Garner, who died during an arrest in 2014. And at a funeral for a slain officer later that year, ranks of police turned their backs when the mayor began to speak. Officers effectively stopped working for two weeks that winter, and arrest rates plummeted.

Despite the tension, the number of homicides in New York City increased just 5 percent last year, well below the average change across the 50 largest cities in the country. The cumulative result of all the controversy over the past several years has been a 35 percent decrease in the number of homicides since 2010.

Then there's Chicago, a city where Mayor Rahm Emanuel has blamed crime rates on closer scrutiny of the police.

"We have allowed our police department to get fetal, and it is having a direct consequence," the mayor said at the meeting in October. "They don't want to be a news story themselves. They don't want their career ended early, and it's having an impact."

Mistrust of Chicago's police has been elevated since an officer fatally shot Laquan McDonald 16 times in October 2014. A grand jury indicted the officer, Jason Van Dyke, on charges including murder in the first degree in November, more than a year after the shooting. Emanuel dismissed his superintendent of police, and protesters paraded along the city's ritziest avenues, calling on the mayor to resign.

The number of homicides increased by 15 percent last year in Chicago. That change is less than the average change across the country's 50 largest cities. The tally last year, 468, was nothing out of the ordinary for Chicago, where no fewer than 500 homicides were committed in 2012.

Heroin

Another theory that Comey cited for the increase in homicides is the expanding popularity of heroin. Possibly, as more people become addicted to the drug, dealers are moving their operations into new territory, leading to disputes and violence.

Richard Rosenfeld, a criminologist at the University of Missouri at St. Louis, points out a problem with this explanation.

Data from the Centers for Disease Control and Prevention shows that more people are dying from using too much heroin, an indicator of Americans' newfound taste for the drug. Yet that increase began in 2010, when the number of homicides was declining in many major cities.

Heroin might account for the greater number of homicides in a city such as Indianapolis, where the count increased by 33 percent between 2011 and 2012 and has worsened since then. Police there have said a dispute over drugs was the motivation in many of those cases.

It isn't clear whether heroin is part of the explanation for the increase in homicides this past year in other cities.

"You'd have to account for at least three years, maybe even a longer gap," Rosenfeld said.

Stephens, of the Major Cities Chiefs Association, ticked off a list of other theories for the increase in violence. Perhaps relaxed gun laws in some states are making firearms more widely available, and more arguments are being settled with lethal weapons as a result.

Stephens also noted that authorities are locking up fewer people in prison, and perhaps more dangerous criminals were on the street last year.

Federal data, however, suggest that the reduction in the incarcerated population over the past several years is mainly a consequence of decreasing admissions, rather than a change in the number of prisoners released annually, which has also declined. In 2014, just 582,000 prisoners were let go from state and federal prisons, compared with 683,000 in 2008.

Additionally, both those explanations are complicated by the absence of any regional pattern in the data. There were more killings in Nashville, but the total in Memphis declined by 1 percent. The number of homicides increased 25 percent in Houston, but decreased 9 percent in San Antonio. There were seven fewer homicides last year than in 2014 in Fresno, Calif., a decline of 15 percent. Meanwhile, up Highway 99 in Sacramento, there were 43 killings last year, an increase of 54 percent.

"Everything is basically anecdotal," Stephens said. "There's not a clear national picture that I've been able to discern of what might be contributing to the changes that we’ve seen in so many cities."

‘Data points’

Wonkblog collected the 2015 homicide data from individual police departments or from news sources that cited those departments. Homicide data for earlier years came from the FBI.

The FBI excludes certain homicides in which authorities determine that the killer did not break the law. Those rare "justifiable homicides" might include acts of self-defense or some killings by police officers. While some police departments do include those incidents in their homicide counts, Wonkblog excluded them from the figures presented in this article.

Justifiable homicides couldn't be removed from the data for seven cities: Colorado Springs; Columbus; Ohio; Los Angeles; Miami; Minneapolis; Seattle; and Wichita. As a result, the counts for those cities might be biased upward relative to previous years.

On the other hand, it sometimes requires coroners and detectives several weeks to determine whether a death is a homicide. Homicides committed near the end of last year might not be included in these totals if investigators haven't yet identified them as suspicious.

While the FBI's official crime data for 2015 won't be reported for several months, the agency recently released preliminary counts for the first half of the year. That data revealed a divergence between large cities and smaller towns. In jurisdictions with at least 1 million people, the number of homicides increased 10.8 percent during the first six months of 2015, and by 12.4 percent in cities with between 500,000 and 1 million residents. Nationally, though, the figure was only 6.2 percent.

Data on other crimes can provide a more complete picture of public safety. In a preliminary analysis based on figures from 19 cities, Chettiar's colleagues at New York University projected that the overall rate of crime declined 5.5 percent last year.

The fact that there has been little change in the number of violent crimes other than homicide suggests that the officers haven't stopped doing their jobs.

At the same time, if heroin addicts are committing more crimes to pay for their drugs, criminologists would expect an increase in property crime, not continuing declines.

The data on homicides does not conclusively rule out a Ferguson effect or heroin consumption as factors in the overall increase in the number of homicides. Both may have contributed to the violence, along with other factors that researchers haven't yet identified.

Experts on crime understand little about what causes fluctuations in violence and lawbreaking over time. That this past year's increase remains mostly a mystery shouldn't be a surprise, since as Wonkblog has previously reported, the much larger decline since 1991 is still largely unexplained.

"We need to figure out what’s happening and deal with it now. I refuse to wait," Comey, the FBI director, said in October. "These aren’t data points. These are lives."