Study findings challenge claims of systemic racism among police
In his presidential nomination acceptance speech, Democrat Joe Biden cited “systemic racism” as an established fact, despite that evidence does not support that sweeping notion — most significantly, not in policing. Such loose talk is incendiary.
In July, the same talk raged inside The Wall Street Journal newspaper, eventually spilling into public view. Almost 300 members of the News division signed a letter to Journal CEO Almar Latour alleging “lack of fact-checking and transparency and apparent disregard of evidence” in the Opinion division, primarily concerning an early June op-ed by Heather Mac Donald titled, "The Myth of Systemic Police Racism."
In the interest of full disclosure, The Journal’s Opinion section has published several of my opinion columns.
Each word in the headline of Mac Donald’s op-ed is important. The title does not call police racism a myth, just systemic police racism. It does not say there are no racial differences, or disparities, in policing.
The headline frames assertively, but fairly and accurately, the column’s central statement that “a solid body of evidence finds no structural bias.”
The News department accused Mac Donald of “cherry picking” a study by Harvard economist Roland G. Fryer Jr. by referencing only Fryer’s finding of no racial disparities, much less racism, in fatal officer-involved shootings (FOIS) while making no mention of Fryer’s finding of substantial racial disparities in police use of non-lethal force.
Well, in mid-June, Opinion ran an op-ed by Fryer, giving him an opportunity to outline his findings on FOIS versus use of non-lethal force and to say, “People who invoke our work to argue that systemic racism is a myth conveniently ignore these (non-lethal force) statistics.”
Yet, in his next sentence, Fryer said, “Racism may explain the findings, but statistical evidence doesn’t prove it. As economists, we don’t get to label unexplained racial disparities ‘racism.’”
Sure sounds like Mac Donald, whom Fryer had just criticized.
The News department also complained that Mac Donald mischaracterized a 2019 study (Johnson and Cesario) which found that minority victims were not more likely to have been shot by white officers.
In April 2020, the authors reconfirmed their findings but corrected a sentence in the study’s “significance statement,” changing “White officers are not more likely to shoot minority civilians…” to say a white officer’s victim, “was not more likely to be of a racial minority.” The change was necessary to recognize that “likelihood,” or probability, of being shot must include data about shots and “non-shots,” just as coin toss probability requires data for both heads and tails, and, since the study’s database was comprised only of fatal “shots” (not even including non-fatal shots) they couldn’t speak to probability of possible FOIS outcomes, but only about the characteristics of actual FOIS outcomes.
Nevertheless, within this more accurate and limited portrayal of their findings, the authors did not find racial bias. Despite this reality, in July, the authors “retracted” their study (whatever that means), because “our work has continued to be cited as providing support for the idea that there are no racial biases in fatal shootings, or in policing in general.”
Except that, as a study that did not find racial bias in FOIS outcomes, inevitably, their work supports previous studies (including Fryer’s) not finding racial biases in FOIS.
Now, what did Johnson and Cesario find? They confirmed previous studies establishing that FOIS victim race varies with crime committed by that race: “as the proportion of violent crime committed by Black civilians increased, a person fatally shot was more likely to be Black…” The authors found this factor to be the predominant predictor, “explaining 44% of the variance in the race of a person fatally shot.”
This is significant: logically, to the extent that something other than racial bias explains racial disparities in FOIS, it lessens the likelihood that, or at least the degree to which, racial bias may explain it.
Contrary to The Journal News division allegations about the Opinion section, it is the signers of the News division’s letter who have misread and misinterpreted the Mac Donald column and the studies cited therein, whose authors have not helped with their confusing and contradictory statements about their own work and the overall issue.
What to do about bad policing? As the subtitle of Mac Donald’s column says, “Hold officers accountable who use excessive force.”
This is not just a semantic or technical debate.
Discrediting Mac Donald’s argument that systemic police racism has not been proven suggests that it has been proven. This inflames anti-police emotions, endangering everyone.
Fryer worries that aggressive pattern-and-practice investigations of possible bias in police departments after major incidents can lead to de-policing and “a stark increase in crime.” He cites the example of Chicago, where de-policing mainly endangers Black citizens.
Delegitimizing the police endangers all citizens, as has become clear since early June as violent crime rates have soared in major U.S. urban centers, and protests have become increasingly violent, with civilians killing civilians in Kenosha, Wisconsin and Portland, Oregon.
Mac Donald highlighted the danger to police, saying “The false narrative of systemic police bias resulted in targeted killings of officers during the Obama presidency.” Her reference is to the five police officers slain in Dallas in 2015, and she warns, “The pattern may be repeating itself.”
It is worth remembering what then President Obama — though also discussing the greater likelihood of Blacks experiencing police confrontations — said at the memorial service for the Dallas officers, “We know that the overwhelming majority of police officers do an incredible job fairly and professionally.”
Red Jahncke (Twitter: @RedJahncke) is president of The Townsend Group Intl. LLC, a Connecticut business consulting firm.
Comment threads are monitored for 48 hours after publication and then closed.