The theory — one of several that criminologist Richard Rosenfeld presents in the paper — suggests that, after a number of widely discussed law-enforcement killings of young black men during the past couple of years, residents of predominately black and disadvantaged urban neighborhoods further lost confidence in the police.
A loss of trust could have made residents of those places less likely to share information with law enforcement about dangerous criminals. With a newfound sense of impunity, these criminals might have begun committing even more crimes. And threatened by the violence, neighbors might have armed themselves instead of going to the police for protection, the theory suggests.
"When persons do not trust the police to act on their behalf and to treat them fairly and with respect, they ... become more likely to take matters into their own hands," Rosenfeld writes. "Disputes are settled informally and often violently."
Relying in part on data collected by The Washington Post earlier this year, the report documents a 17 percent increase in homicides in the country's largest cities last year. The Justice Department commissioned the report to suggest an agenda for research on the violence, but it notes that the conclusions do not reflect the views of the agency or the Obama administration.
Rosenfeld was an early critic of the notion that there might be some connection between the recent police killings and the increased crime — a theory that's been called "the Ferguson effect." President Obama rejected another version of the theory last year, saying there was no evidence that increases in urban crime were a result of police's pulling back on their duties.
Rosenfeld, though, now believes that such a connection is the most likely reason for the increases in crime, although the evidence is scant and the details unclear.
"My views did change, looking at more data," said Rosenfeld of the University of Missouri at St. Louis, noting that there are several possible theories to explain the outbreak of violence. "I don't subscribe to a particular explanation yet."
There is some evidence that national attention on police shootings of civilians worsened African Americans' views of law enforcement.
Immediately after Michael Brown's death in Ferguson, Mo., in 2014, the Pew Research Center asked people whether they trusted the police to treat people equally, regardless of race. Forty-six percent of black respondents told Pew that they had "very little" confidence in the police to do so. When Pew asked the same question in 2009, just 34 percent of black respondents were so pessimistic.
Gallup found a similar change in just one year. In a 2015 poll, 30 percent of black respondents said they had "a great deal" or "quite a lot" of confidence in police, a six-point decline from the year before. (There was a three-point decrease among white respondents, to 57 percent).
It isn't clear whether this decrease in confidence was more pronounced in cities where homicide increased or whether it was the cause of the increases.
Still, "a subtle shift in the perception of law enforcement, if the city is large enough, can lead to what looks like a significant increase in the body count," said Phillip Atiba Goff, a psychologist at UCLA. "It happens very quickly."
No good explanation
Confirming findings in The Post's analysis, Rosenfeld's report determined that the worst violence was concentrated in a few cities. Ten large cities accounted for two-thirds of the overall increase. The greatest absolute increases were in Baltimore, Chicago, Houston and Milwaukee.
Fortunately, urban violence did not return to the extreme levels of the recent past, despite the increase. There were fewer homicides in the 50 largest U.S. cities last year than there were in the same cities as recently as 2008.
All the same, the outbreak of violence is worrying, and it is all the more so because criminologists still have not come up with a good explanation for the bloodshed.
Some observers — including FBI Director James B. Comey — have suggested that last year's increase in homicides could be connected to increased public scrutiny of the police, after the deaths of several young black men at the hands of law enforcement.
According to this argument, the police have become hesitant to do their work — averse to speaking with people on the street when it is not necessary or wary of using force to apprehend criminals for fear something could go wrong — resulting in police who are less effective at preventing crime.
"There’s a perception that police are less likely to do the marginal additional policing that suppresses crime — the getting out of your car at 2 in the morning and saying to a group of guys, ‘Hey, what are you doing here?’ " Comey said last month.
Rosenfeld is skeptical of this theory. "It's possible, but we have no evidence," he said.
He explained that outside a couple of cities — Baltimore, for example, and possibly Chicago — the rate of arrests has not declined, suggesting that police are working just as hard as before.
He thinks, though, that there might be some kind of connection between recent police killings and the increase in homicides. For Rosenfeld, such a link appears to be the only way to explain the abrupt shift in homicide rates last year.
He also considered an increase in the number of people being released from prison and an increase in the demand for narcotics and opioids. While the population behind bars has declined, and although the number of deaths from heroin overdoses has increased, both of those trends began several years ago, so it is unclear why they would result in more homicide only now.
So far, there is no direct evidence connecting last year's increase in homicides to the perceived legitimacy of the police among black Americans, but there is some data to support the idea.
Using The Post's data, along with a survey by the Major Cities Chiefs Association, Rosenfeld found that the average black population of the 10 cities with the severest increases was more than twice that of the rest of the cities.
There is evidence that, where the police are seen as legitimate, they get more help from bystanders in fighting crime. A study in New York City in 2002 and 2004 found that people who had more favorable views of the police were also more likely to say that they would report a crime in progress or help police locate a suspect.
Among the city's racial and ethnic minorities, those who viewed the police more favorably also said they would be more willing to work with other civilians in their community to control crime — for example, by patrolling the streets with other volunteers.
A 2012 study found a similar effect, focusing specifically on residents of Chicago who were known firearms offenders. Just 31 percent of them agreed with the statement: "Most police treat people with respect." Although all participants in the study would have had encounters with the police, those with more positive views of the law were less likely to say they carried guns outside their homes, an act that would have been illegal under Chicago's laws at the time. Put simply, confidence in police was correlated with law-abiding behavior.
"When people do not perceive the police to be legitimate in their practices and actions, crime rates tend to increase," Rosenfeld said. "They don’t believe the police are there to protect them, and that leads to increases in retaliatory killings."
'We simply don't know'
Many African Americans, based on long experience, have always felt that brutality and racism among the police are widespread. Whether the deaths of Brown and others would change that perception enough to result in large increases in violent crime is uncertain.
Like Pew's data, Gallup's polling reveals a broad decline in public confidence in the police. Just 23 percent of respondents of color rated the honesty and ethics of the police at least "high" in the organization's poll in 2014. That share increased last year, but remained at its lowest level in almost a decade.
"Even though levels of mistrust were high in the last several years, especially among black Americans, they've gotten even higher," said David Kennedy of the John Jay College of Criminal Justice.
Still, these national data reveal little about how the police are seen in the specific places where most of the murders are being committed. "We don't have a local sense of what’s going on," said Goff of UCLA.
For now, the idea that a crisis of legitimacy in American law enforcement could explain last year's increase in urban homicide remains no more than a provocative hypothesis.
"We simply don’t know what’s happening," Kennedy said.