In the Two Minute Drill, we explain complex issues in politics in 500 words or less (roughly the amount of words it takes the average adult two minutes to read on a monitor). Politics just isn’t always that complicated. Without the fluff and partisan bias, even the most complex of our political differences can be explained succinctly. This is The Two Minute Drill for July 15, 2016.
The Issue: Bias in the Use of Police Force
A recent working paper by Harvard economist Roland Fryer on the topic of racial bias in police use of force and police shooting gained substantial media attention after it purported to show that there was no evidence of racial bias in police shootings. A write-up of the study became the top viewed article on The New York Times website.
The Times reported: “A new study confirms that black men and women are treated differently in the hands of law enforcement. They are more likely to be touched, handcuffed, pushed to the ground or pepper-sprayed by a police officer, even after accounting for how, where and when they encounter the police. But when it comes to the most lethal form of force – police shootings – the study finds no racial bias.”
Mr. Fryer told The New York Times that the finding was “the most surprising result of my career.” Which begs the question, is this “finding” correct? Does a new study really show that there’s no racial bias in police shootings?
The Explanation: Roland Fryer Is Demonstrably Wrong
Roland Fryer is a distinguished Harvard economist. But he is demonstrably wrong and he has communicated the results of his study to news media in a way that is highly misleading.
First, Fryer’s study suffers from significant theoretical problems. If you take a look at the same Houston police shooting dataset that Fryer used for the years 2005-2015, you will find that black people were over 5 times as likely to be shot relative to whites. So what explains the studies results? Well, Fryer was not comparing rates of police shootings by race. Instead, his research asked whether these racial differences were the result of “racial bias” rather than “statistical discrimination.”
These terms have specific definitions in economic theory. Statistical discrimination occurs when people are treated differently based on racial stereotypes that “truly” reflect the average behavior of the group. For example, if a city’s black pedestrians are 50% more likely to possess drugs than white pedestrians, and police officers are 50% more likely to stop-and-frisk black pedestrians, economic theory would hold that this discrimination is rational. Racial bias, on the other hand, occurs when statistical discrimination gets out of hand – that is, when police exceed the rational level of discrimination and collectively punish people based on the “purported” average behavior of the group. If cops were to stop-and-frisk black pedestrians at a rate that disproportionally exceeded their likelihood of drug possession – let’s say, for example, 65% in the scenario above – that would represent statistical discrimination which turns into racial bias.
The use of stop-and-frisk in the above examples is done deliberately because this is the framework that has been used to study and interpret these practices (along with studies involving traffic stops). But it is an inappropriate framework for studying police shootings. Underpinning the idea of statistical discrimination is the notion that police engage int hat behavior to maximize the number of arrests while expending the fewest resources. In other words, the racial stereotypes – regardless of how abhorrent you may find them – could result in the officers acting in the most cost-efficient, rational manner. However, that cannot be said about police shootings. It would be absurd to even suggest that officers are trying to rationally maximize the number of police shootings. The basic theory underpinning Fryer’s study, therefore, is completely inapplicable.
Second, Fryer’s study suffers from fatal methodological errors. One issue has to do with the target population of the study. In a typical research scenario, the researchers confine their variables to a previously defined population where each individual is at risk of a particular outcome. For example, pedestrians who are the subject of a stop-and-frisk can have one of two outcomes: they can be arrested, or they can be sent on their way. Fryer, though, has chosen to focus on a fictitious, and wholly unrelated, population – namely, people who are shot by police and people who are arrested. In doing so, Fryer has introduced a multitude of variables into his study that, in all likelihood, could not be controlled for statistically.
A bigger problem, though, is the size of the population analyzed. Fryer looked at 1,332 shootings between 2000 and 2015. But these shootings came from police reports from just one city: Houston, Texas. The actual data – which, again, shows that black people were over 5 times as likely to be shot relative to whites – calls into question Fryer’s results. But even if we take Fryer at his word about the city of Houston, it is doubtful the city is representative of the country.
Consider Chicago, for instance, where a recent review found a disproportionate 118 black males (44 of them fatal) were involved in the 150 shootings recorded since 2010. Chicago is not an aberration; these trends hold nationally. In fact, most studies have shown a greater propensity for shooting black civilians relative to whites. In addition, a recent analysis of national data showed wide variation in racial disparities for police shooting rates between counties, and “found significant bias in the killing of unarmed black Americans relative to unarmed white Americans.”
So, while there are demonstrated problems with the quality of police shootings data, there is still plenty of evidence to support racial bias in the use of force against black people in the United States.
Featured Image: Phil Roeder on Flickr (via creative commons)