In November, Donald Trump tweeted out an image of 2015 homicide statistics from the “Crime Statistics Bureau” of San Francisco that purported to show that the majority of black homicides were committed by other black Americans. Black-on-black homicides were shown to be 97 percent of all black homicides, while white-on-white homicides were a mere 16 percent of all white homicides.
If that difference sounds too incredible to believe, that’s because it is.
These statistics have already been roundly and rightly taken apart. The Crime Statistics Bureau doesn’t exist, and we don’t have public homicide statistics by race for either San Francisco or the nation for 2015. This isn’t a controversial scholarly question of how you slice the data or whether one arcane statistical test or another is the right tool: the data simply don’t exist.
Trump’s numbers are wrong. But what is most revealing is how the numbers are wrong.
The most glaring error isn’t the black-on-black homicide statistic, which in 2014 (the last year for which we actually have national data) was 90 percent of black homicides. The most glaring error is the white-on-white homicide statistic. In 2014, 82 percent of white homicides were committed by other white people. That means that while black-on-black homicide is overstated by 7 percentage points, white-on-white homicide is understated by a whopping 66 percentage points.
The myth of extreme black criminality looms large in American discussions of crime, supported by misrepresentations like these. This myth persists despite recent research demonstrating that racial differences in criminal offending are limited, particularly for serious, juvenile offenders, who often feature conspicuously in discussions of violent crime and homicide. When socioeconomic differences are controlled for, the case for black hypercriminality gets even weaker.
Despite the research, violence in black communities is frequently brought up to derail policy conversations about the justice system and its effect on black communities. In fact, this routine is so common that we saw it at the exact same time last year. When asked about police violence in black communities, Rudy Giuliani noted that 93 percent of blacks are killed by other blacks. This time, white-on-white violence was not misreported but simply left unsaid, once again with the implication that black Americans are uniquely violent.
Both Trump and Giuliani’s remarks draw on the myth of black hypercriminality to short-circuit policy conversations. Trump’s tweet came on the heels of a confrontation with a Black Lives Matter protester at one of his rallies, while Giuliani employed the trope to defend existing police tactics that were being sharply criticized. In both cases, high levels of black-on-black homicide were cited as if they were unique to black communities, when in fact, most violence tends to occur between people of the same race.
Trump’s numbers are dead wrong. However, the fiction of black hypercriminality they support is even more grossly inaccurate.
This article was originally published by the Urban Institute.
