Kent Osband, author of “Pandora’s Risk: Uncertainty at the Core of Finance”, addresses issues raised by police shootings and resulting protests in Ferguson, MO, Madison, WI, and Brooklyn, NY by asking “what is the statistical evidence of race-based bias in police shootings of black civilians?”
Using the FBI’s Uniform Crime Statistics reports (which disclose that 1/3 of all police shootings involve black citizens) and extrapolating, Osband calculates that the total number of civilians killed by police is not the FBI’s reported 400/year, but, instead, is closer to 800/year after adjusting for reporting inadequacies, which means that nearly 270 African-Americans are likely killed annually by police.
Over the past ten years, 51 officers have been killed annually in the line of duty. On average, 24 of those 51 officers are killed by black civilians. Adjusting for population numbers, African-Americans are 5.6 times more likely than non-blacks to kill an officer. Considering the relative numbers of police vs. blacks, police are six times more likely to be killed by a black civilian than a black civilian is likely to be killed by police. On the other hand, police respond far more defensively to blacks (1:6 homicide rate) than to non-blacks (1:22 homicide rate), so African-Americans can correctly conclude that police are three to four times more likely to pull the trigger against a black perceived as a dangerous threat than a non-black.
- Statistically, police have more to fear from the public than the public has to fear from police.
- While blacks account for 14% of the US population, they account for 47% of police officer killings.
- In a random sampling, blacks are 5.6 times more likely than non-black civilians to kill a police officer.
- Police are 3-4 times more likely to shoot a black American than a non-black American.
Given the statistical likelihood of a fatal encounter in a high stress encounter with limited information, many blacks and police officers are right to be genuinely fearful of each other.