Heather Mac Donald is a fellow at the Manhattan Institute and a contributing editor of City Journal. She is the author of The War on Cops: How the New Attack on Law and Order Makes Everyone Less Safe.
Some statistics on our discussion: Black are men nearly three times as likely to die from police use of force, as judged by how they are represented in the population, however Heather makes the point that it is more valid to calculate the proportion not by representation in the whole population, but by representation in the population that encounters the police, such as being arrested.
I believe that this point is weak because if the hypothesis is that black men are more likely to die because of police racism, it is not surprising that they are also more likely to be arrested in instances where a white man would not be.
Research shows that although blacks are arrested far more often for drug use, they report drug use at almost exactly the same levels. Heather contends that this is because blacks falsely under-report their levels of drug use. This is not impossible, but it seems surprising that blacks falsely under-report their drug use to exactly the right degree to make them appear to have the same level of drug use as whites.
Heather is right that the statistics strongly suggest that members of some minorities are disproportionally responsible for serious violent crime, and is also right that, despite that, most members of minority communities have no involvement in crime.
Criminologists agree that statistics for serious crimes are generally accurate, and the more serious they are, the more accurate the statistics – a burglary might go unreported, a murder hardly would. In addition, the more serious the crime, the less scope there is racial bias in dealing with it – it would be startling for a police officer to fail to arrest a murder suspect of any race, although it has happened.
However, for less serious offences, police have wide discretion whether to arrest a suspect or not; naturally the less serious the offence, the less likely an arrest. But statistics indicate that police officers are far more likely to arrest black suspects when the offence is less serious – such as minor drug use – and this effect is magnified the closer you get to the most trivial end of the scale of offences.
This is seen most clearly in the arrest rates for very minor traffic offences, and also in the huge disparity in the arrest rates for black and white women.
When seen in this light, it seems less reasonable to measure fatal interactions with police as a proportion of arrest rates, rather than population distribution.