University of Missouri at St. Louis criminologist Richard Rosenfeld has had "second thoughts." Like many academic criminologists, he had pooh-poohed charges that skyrocketing murder rates in many cities in 2015 and 2016 result from a "Ferguson effect" -- a skittering back from proactive policing for fear of accusations of racism like those that followed the death of Michael Brown in Ferguson, Missouri, in August 2014.
Now, after looking over 2015 data from 56 large cities, he tells The Guardian that he has changed his mind. Homicides in those cities were up 17 percent from 2014. And 10 cities, all with large black populations, saw homicides up 33 percent on average.
"These aren't flukes or blips, this is a real increase," Rosenfeld told reporter Lois Beckett. "The only explanation that gets the timing right is a version of the Ferguson effect."
Rosenfeld thus parts company with the liberal Brennan Center, whose analysts argued that the 2015 homicide increase in large cities was not a "national pandemic." He parts company also with FiveThirtyEight analyst Carl Bialik, who dismissed a 16 percent homicide increase in 59 of the 60 largest cities in 2014 and 2015 as "a less dire picture than the one painted by reports in several large media outlets."
But a 16 or 17 percent increase in homicides in major cities that account for a large share of the national murder toll is, in historical perspective, very dire indeed. The most accurate word is "unprecedented." The only double-digit increases in national murder statistics going back to 1960 are 13 percent (in 1968), 11 percent (in 1966, 1967 and 1971) and 10 percent (in 1979).
As anyone familiar with the workings of compound interest might guess, such increases rapidly added up. The total number of homicides nationally more than doubled between 1966 and 1979. The number peaked in 1991.
During those years, most academic criminologists argued that high rates of violent crime resulted from economic distress and -- noting that nearly half of murders were committed by blacks -- from the endemic racism in American society. Today the Brennan Center echoes this analysis: "Economic deterioration of those cities could be a contributor to murder increases."
Political scientist James Wilson and maverick criminologist George Kelling dissented from this view. In their 1982 Atlantic article "Broken Windows" they argued that proactive policing and elimination of signs of disorder (like broken windows) could sharply reduce crime rates.
In the 1990s New York Mayor Rudy Giuliani and Police Commissioner Bill Bratton put the "broken windows" theory into effect. Their proactive policing tactics were continued by Mayor Michael Bloomberg and widely imitated and adapted around the country.
The result was that homicides in New York were reduced from 2,445 in 1990 to 328 in 2014. Nationally, the number of murders declined 42 percent from 1991 to 2014.
The definitive chronicler of proactive policing, the Manhattan Institute's Heather Mac Donald, spotlighted the Ferguson effect in a Wall Street Journal article in May 2015. She noted that arrests were sharply down in cities such as St. Louis and Baltimore because the "incessant drumbeat against the police" across the country had "officers scale back on proactive policing under the onslaught of anti-cop rhetoric."
Those encouraging such rhetoric include President Obama and his first attorney general, Eric Holder -- even though an intensive Justice Department investigation of Brown's killing in Ferguson cleared the officer involved and made clear that charges that Brown had put up his hands and surrendered were baseless.
Obama has since said that "there's no data to support" a Ferguson effect. That puts him at odds with his appointee FBI Director James Comey, who says that his conversations with police officials around the country convinced him there are "marginal pullbacks by lots and lots of police officers." It also put Obama at odds with Rosenfeld, who has found clear evidence of "de-policing" in Baltimore and Chicago, where homicides have spiked.
The charge of cherry-picking data and misleading rhetoric can more justifiably be leveled against administration officials and mainstream media, who, after the Ferguson killing, created the impression of a rising epidemic of racist police officers shooting innocent blacks. The few such cases have received prompt and stern attention from local law enforcement.
Black Americans were the primary victims of the huge crime increase starting in the late 1960s, and they will be the primary victims again if the Ferguson effect continues to result in more homicides. Can't we prevent this awful history from repeating itself?