A federal report released Wednesday examines FBI crime data in big cities, considering two possible explanations for the “sudden and unforeseen” spike in homicides nationwide.
According to the National Institute of Justice (NIJ) study, which focuses on cities with a population of 250,000 or more, preliminary evidence suggests that expansions in illegal drug markets have driven the considerable increase of homicide rates among whites. In 2015 and 2016, drug-related homicides increased to a greater extent than other types of homicide.
The second hypothesis authors consider is the so-called Ferguson effect, which resulted in “de-policing, compromised police legitimacy, or both.”
NIJ stipulates that “current evidence that links de-policing to the homicide rise is mixed, at best,” and that it remains an “open research question”– since arrests-offense ratios and arrest clearance rates had been declining for years, as were homicide rates, before the recent spike.
Neither explanation is meant to stand alone; in particular, growing tensions between African Americans and police do not account for the abrupt increase in homicides among whites beginning in 2013.
A look at the numbers
Nationally, 2014-2015 saw the largest percentage rise in the homicide rate since 1968, according FBI data; an increase of 11.4% over the previous year, or from 4.4 to 4.9 homicides per 100,000 people.
National homicide rates continued to increase by 8.2% in 2015-2016, and 10.8% in big cities. In 2015, Cleveland and Nashville saw the biggest absolute homicide increases, while Austin and Chicago topped the chart in 2016.
Only two cities bucked this upward trend: Miami and Tucson both experienced a decline in homicides over the two-year period.
Despite the recent incline, the homicide rate in 2016 was still 35.4% lower than it was in 1995. “Even at the elevated rates of increase in 2015 and 2016,” note the authors, “it would take about five more years for the U.S. national and big city homicide rates to return to the levels of the early 1990s.”
‘Sudden and unforeseen’
In the search for external ‘shocks’ that could explain the rapid and unexpected nature of the recent homicide increase, NIJ finds two plausible candidates, both of which have parallels in contemporary U.S. history: the opioid epidemic, echoing the crack-driven homicide escalation of the 1980s–and anger over police brutality and fatal use-of-force, such as precipitated the civil unrest of the 1960s (and corresponding “crisis of institutional legitimacy,” according to some analysts).
Researchers need better data to measure the affects of the opioid crisis, according to the report. Drug arrests are not a reliable indicator, since they are “a product of both police enforcement and criminal conduct,” and since “policymakers and law enforcement officials alike have viewed the heroin and synthetic opioid epidemic as more of a public health than a criminal justice problem.”
There is some data to suggest a link between heroin and increased homicides: arrests for heroin or cocaine were falling between 2010 and 2013, but heroin use was on the rise. Arrests then rose in 2014 and 2015, coinciding with the accelerating homicide rate.
In future studies, authors recommend using data that is disaggregated by race, in order to “determine whether, as expected, whites have entered local drug markets in greater numbers over time as both buyers and sellers.”
Police activity and community “reservoir of discontent”
When considering the influence on homicide rates of police enforcement, legitimacy, and community alienation, reliable indicators are also hard to chase down.
Arrest rates not only measure crime, but police activity, as noted above. “Arrests fell in Baltimore in 2015 after the Freddie Gray incident, and in Chicago in 2016 after the delayed release of a video of a controversial police shooting there,” according to the report. In both cities, the reduction in arrests was then followed by an increase in homicides.
However, researchers say that it “remains to be seen whether comparable decreases in arrests preceded increases in homicide elsewhere.” The link between homicide and arrests/de-policing needs to be further examined at the neighborhood level, since homicide rates vary substantially across neighborhoods.
Calculating police legitimacy and community alienation is “onerous,” and largely measured through opinion surveys– “for the time being,” say the authors, “it appears that strategic case studies and one- or two-time snapshot surveys will have to suffice.”
There are two empirical indicators that can be measured, however, if police departments are willing to release the data: calls for police service, and complaints against the police.
“If the community alienation hypothesis is correct, investigators should expect to observe a reduction in calls for service and an increase in complaints in cities where controversial use-of-force incidents and outbreaks of community unrest have occurred, particularly in African-American communities. Increases in homicide…should be greater in those cities and communities than in others.”
The dark figure of homicide
Finally, and critically, “it is important not to overstate the precision of these figures,” say researchers. “Like all UCR crime data, they are based on the classification of homicide events by local law enforcement agencies, and crime classification criteria and procedures can differ across agencies or within the same agency over time.”
Homicides with unknown circumstances are omitted from the FBI’s Supplemental Homicide Report– and “the circumstances of fully 39.9 percent of homicides were unknown to law enforcement officials in 2015.”
To locate many of the indicators that identified in the NIJ report, researchers will have to go beyond FBI data and work directly with public health and police sources.