Murder rates rose significantly in 25 of the nation’s 100 largest cities last year, according to an analysis by The New York Times of new data compiled from individual police departments.

In the Times analysis, half of the increase came from just seven cities — Baltimore, Chicago, Cleveland, Houston, Milwaukee, Nashville and Washington.

Rosenfeld is a significant figure in the debate over rising homicide rates because he has effectively switched sides. In 2015 he came out against the so-called Ferguson effect and was widely quoted to the effect that is was not real. However, earlier this year he published the paper (linked by the NY Times above) in which he seemed to judge the Ferguson effect the most plausible explanation. Here’s a bit of his paper:

Lack of confidence in the police among African-Americans predates the recent police killings in Ferguson, Cleveland, New York and elsewhere. But it is likely to be activated by such incidents, transforming longstanding latent grievances into an acute legitimacy crisis. If that led to the 2015 homicide increase, we should expect at least four empirical conditions to hold: (1) the increase should be concentrated in cities with large African-American populations, (2) the timing of the increase should correspond closely to controversial incidents of police use of force against African-Americans, (3) confidence in the police should be substantially lower among African-Americans than other groups and (4) the homicide increase should be greater among African-Americans than other groups.

Rosenfeld goes on to say that the first three are easy to prove. On the fourth point, Rosenfeld says there are two definitions of the Ferguson effect, once that involves police pulling back and another that involves criminals feeling empowered. The first of these, he says, will be easy to assess once the FBI releases 2015 crime data in the fall. The alternative definition will be harder to prove and would lack a clear mechanism connecting widespread news of police shootings to criminal activity. Still, Rosenfeld seems to believe some version of the Ferguson effect is likely a factor in the rising homicide rates.

The Times gives Rosenfeld’s argument short shrift, only mentioning one of the two definitions of the Ferguson effect he brings up, downplaying how much space he devotes to it and then quickly dismissing it:

In his study, Dr. Rosenfeld said that rising crime might be linked to less aggressive policing that resulted from protests of high-profile police killings of African-Americans. But he said this hypothesis, a version of the so-called Ferguson effect, which has spurred heated debate among lawmakers and criminologists, must be further evaluated.

There is no consensus on what caused the recent spike, and each city appears to have unique circumstances contributing to the uptick.

Even so, the Times does note the obvious connection, at least in part:

At least three of these cities have also been embroiled in protests after police-involved deaths of black males, like Freddie Gray in Baltimore, Laquan McDonald in Chicago and Tamir Rice in Cleveland.

That leaves out Ferguson which is only a few miles from St. Louis. St. Louis doesn’t get much attention in the Times’ story since it’s not one of the 7 cities identified as driving the increase. However the Times does say almost in passing, “Baltimore had the largest increase — 133 more than 2014 — and the second-highest rate in 2015, after St. Louis, which had 59 homicides per 100,000 residents.” This sentence manages to mix raw numbers and per capita rates, making it a bit hard to follow. If I’m reading that correctly St. Louis had the highest per capita murder rate in 2015. That seems like a rather significant fact to leave dangling at the end of a sentence without comment, especially when the Ferguson effect is an issue in the story.

Here’s the graphic created by the NY Times showing how St. Louis and Baltimore stood out: