Study: There Is No 'Systematic Ferguson Effect'
After a white police officer shot a young unarmed African-American in Ferguson, Missouri in 2014, pundits began talking of a so-called “Ferguson Effect.” The theory goes: that after the shooting of Michael Brown, and other high profile cases around the country, crime rates have surged in major U.S. cities. A recent study from researchers in Arizona, Colorado, and South Carolina, though, says the discourse around the Ferguson Effect is “long on anecdotes and short on data.” The study looked at crime data from 81 large U.S. cities, including Cleveland and Cincinnati, seeking signals that crime is ballooning. Ideastream’s Tony Ganzer spoke via Skype to Scott Wolfe, a co-author of the study. He’s assistant professor of Criminology and Criminal Justice at the University of South Carolina.
WOLFE: “Systematically across the United States there doesn’t appear to be a ‘Ferguson Effect’ on crime rates. But there does appear to be somewhat of a, if there is a ‘Ferguson Effect,’ it’s constrained to a certain number, a handful of cities that have similar structural characteristics in common with one another.”
GANZER: “I’ll just read a line here, you say ‘any Ferguson Effect is constrained largely to cities with historically high levels of violence, a large composition of black residents, and socioeconomic disadvantages.’ How does Cleveland fit into that?”
WOLFE: “It fits right into that group. And so Cleveland, another city within Ohio Cincinnati, would be two of those cities that experienced an uptick in violent crime, but specifically an uptick in homicide rates in the post-Ferguson era. And so corresponding right around the same time as Michael Brown’s death in Ferguson, in August 2014, these cities, cities like Cleveland, witnessed a redirection of the homicide rate within those cities.”
GANZER: “How big an increase are we talking about? Is this statistically significant?”
WOLFE: “Very good question. Statistically significant at least within those cities, that group of cities, that seemed to witness some type of ‘Ferguson Effect.’ It is statistically significant. But the key within that finding is the nuance. Substantively it appears to be very, very small in magnitude. And so on average you would expect across a handful of cities it would take almost two years across all those cities to witness about a one unit increase in homicide per 100,000. So we’re talking about a statistically significant effect, but substantively nothing to really raise alarm bells over.”
GANZER: “I didn’t see Columbus on your list, is that right?”
WOLFE: “That’s correct, yeah, unfortunately that was one of…we had 81 out of the 105 largest with populations of 200,000 or more, and we tried, we tried a lot to get data from all of those cities, and Columbus, unfortunately, happened to be one that we couldn’t get the data for.”
GANZER: “So one of the constraints you had was police weren’t giving you the data?”
WOLFE: “Yeah, not so much, I don’t want to fault them. The problem we encountered is, we’re relying on essentially uniform crime report data from the FBI, and so those data are made available by the FBI to the public, but there’s about a year, sometimes a two-year lag in when those data become available. So everything up to 2013 is publically available, and then most of 2014 for some cities, but then we stop. So what we had to do at that point was basically cold-call contacts within agencies. You know, crime in the United States is one of the most significant social problems, public health problems we face, and we have a big debate surrounding this ‘Ferguson Effect’ and we can’t answer it fully in a good amount of time.”
GANZER: “Is that where we leave it then, that we can’t answer it? Or do you think it’s debunked based on what you’ve seen?”
WOLFE: “We’re very confident in the cities that we have, there’s no systematic differences between those that are included and those that weren’t able to get complete data for. From a debunked standpoint, I would say that the overarching hypothesis that there is a nation-wide crime wave stemming from the ‘Ferguson Effect’ is definitely debunked. The important thing to come away with it is that doesn’t necessarily mean there are no Ferguson effects, though, right? Certain cities in the United States like Baltimore, St. Louis, even Cleveland, did experience upticks in homicide. Substantively is where we need to come down on whether these are big enough increases to warrant enough public attention that may be instrumental in guiding public policy.”