“Data collected by the Ferguson Police Department from 2012 to 2014 shows that African-Americans account for 85 percent of vehicle stops, 90 percent of citations, and 93 percent of arrests made by FPD officers, despite comprising only 67 percent of Ferguson’s population.”
Those statistics don’t prove racism, because blacks don’t commit traffic offenses at the same rate as other population groups.
The Bureau of Justice Statistics’ 2011 Police-Public Contact Survey indicates that, nationwide, blacks were 31 percent more likely than whites to be pulled over for a traffic stop.
Ferguson is a black-majority town. If its blacks were pulled over at the same rate as blacks nationally, they’d account for 87.5 percent of traffic stops.
In other words, the numbers actually suggest that Ferguson police may be slightly less likely to pull over black drivers than are their national counterparts. They certainly don’t show that Ferguson is a hotbed of racism.
The Justice report on Ferguson continues, “African-Americans are at least 50 percent more likely to have their cases lead to an arrest warrant, and accounted for 92 percent of cases in which an arrest warrant was issued by the Ferguson Municipal Court in 2013.”
Again, this pretends that a mere difference is evidence of discrimination.
But the report’s statistic doesn’t even look at whether people pay their fine or appear in court — something that makes a big difference in whether to issue a warrant.
Could it be that blacks are more likely to face particularly serious charges?
One of the favored techniques in civil rights law, borrowed from social science, is disparate-impact analysis. The gist of this methodology is to compare the percentage of a particular group in a population with the percentage of those affected by the activity under study. If a wide variance exists between the two figures, that constitutes disparate impact. So, for example, if we want to determine whether, due to race bias, African-Americans are arrested in greater numbers than seems warranted by their population size, one would first establish the percentage of blacks in the population subject to arrest, then the percentage of all arrested persons who are African-American.
The next step—the hard part—is to explain the reasons for the variance, which may or may not involve bias. To continue with the example of arrests and African- Americans: if the proportion of black arrests is much greater than the proportion of the black population, then police could be treating blacks differently because of the color of their skin. It also could be that African-Americans engage in criminal activity out of proportion to their numbers, in which case bias may not be a significant explanation. Needless to say, this issue is both controversial and difficult to resolve. But one thing is clear: it’s wrong to assume bias whenever there is a racially disproportionate impact.
Using 2010 census figures for Ferguson, the Justice Department documents that about two-thirds of the city’s 21,000 residents, or 67 percent, were African-American. Interactions with police over and above the 67 percent figure would be disproportionate, but not necessarily bias-driven. After all, human groups don’t all engage in the same behaviors, and it’s possible that African-Americans commit offenses out of proportion to their numbers. But let’s set that contentious claim to one side for the moment.
When criminologists examine unlawful behaviors, they take into account factors associated with such misconduct. We know, for instance, that males commit more violent crimes than females, and that young persons are more crime-prone than the elderly or children. Consequently, when we look at the criminogenic population in Ferguson, we should account for age and gender. If we single out the young male population of the city using the census parameters for age, we find that blacks make up approximately 80 percent of the population between the ages of 15 and 29, not 67 percent. The median age of black males in Ferguson is roughly 25, whereas it is 46 for white males, according to census data. Thus, 80 percent may be a more relevant figure for measuring disparate treatment by police than 67 percent. And that means that the disparities reported by the Justice Department are 13 percent less than claimed.