[The Harvard sociology cohort of 1995 always turned to Morgan as the quant guy when we needed help with stats class, which was often. So rather than blame my own limitations and laziness, I prefer to entirely and falsely blame Steve for the fact that I still can't really tell you what a poisson regression is and why you would want to use one.]
I had some input in the drafts. One of my points was that the take-away would be an idiotic headlines like this one: "Study: There Has Been No 'Ferguson Effect' in Baltimore."
Citilab never talked to Morgan, which seems odd.
Of course that headline isn't the point of the study. I think the narrative focus of the study should have been centered around April 26, 2015 (the riot) and not early events half a country away (Ferguson). All policing is local. Buried halfway down that Citilab story is a mention of the "Gray Effect." It is a better term. Perhaps because of the double meaning of gray, it can be applied elsewhere in a generic sense.
I'm baffled by many people's attempt to disaggregate a so-called "Ferguson Effect" from local police issues, since I've been arguing this is the same thing. But shorthand terms are only helpful if they have an accepted meaning. And clearly the Ferguson Effect does not. I'm not willing to waste time in a semantic debate or defend a term -- the Ferguson Effect -- that I never liked. So let's call it the Gray Effect. My point is that police matter and that society influences policing, sometimes for the good and -- as last year's spike in homicides portends -- sometimes for the bad. Call it what you will, the effect is real.
Better reporting is done by Baynard Woods and NBC. From the latter:
"I do think we provide some pretty compelling evidence that it is possible for the police to use discretion, to use alternatives to arrest, in a place like Baltimore without influencing the pattern of crime," Morgan said.[Maybe it's minor, but I'll take credit for the subtle addition of "lack of support from the city's leadership," thank you very much. Correction: Steve, ruining my fun as only a quant guy can, says that phrase was in the earliest drafts and had nothing to do with me. --eye roll-- ]
That is why Morgan says the eight months before Gray's death could represent a "sweet spot."
The next part of Morgan's analysis, the Gray period, was much less surprising.
"Everything fell apart," Morgan said.
Crimes of all types, violent and non-violent, spiked, for an overall increase of more than 11 percent. [Ed note: In reported crime.... Homicides doubled, and there is good reason to believe more crime was non-reported. And decreasing arrests will also serve to reduce crime stats without a corresponding reduction in actual crime.] The drop in arrests became much more pronounced, from 19 percent to 30 percent, "consistent with the widely discussed conjecture that the Baltimore police pulled back from some routine policing in response to a perceived lack of support from the city's leadership," the researchers wrote.
From Woods in the Guardian:
“One reasonable interpretation of these entangled effects is that the crime spike in the Gray period could be a Ferguson effect that would have remained dormant had it not been ignited by a localized Gray effect,” the report states. “However, the size and duration of the crime spike is almost certainly attributable to particular features of the unrest.The whole point of the Gray Effect (née Ferguson Effect) is that it is not necessarily centered around the events of Ferguson. Let's the just accept that and move on. It is about media focus and changing political pressures of the past few years.
The study found a decrease in crime in the period after the new police commissioner, Kevin Davis took office, which they dub the “Davis effect.” Davis replaced then-commissioner Anthony Batts, who was fired just after a Fraternal Order of Police report criticizing his handling of the riot came.
The substantive issue is that anti-police movements and protests can affect policing and policing impacts criminals and crime. The events around the riots in Baltimore -- specifically the failure of political leadership and the politically motivated prosecutions of police officer -- were Baltimore's
Public events, media reporting, and political leadership all matter to police officers. And when this process is happening in many different cities, a shorthand label can be useful. When the factors combine to change policing in a negative way -- when police are less proactive and more young black men are killed as a result -- we need to recognize the facts and react accordingly.