D.C.'s Aberrant Homicide Data Since 2012

In general, crime tends to coincide with other types of crime. Property crime coincides with violent crime. Higher rates of assault coincide with higher rates of robberies, murders, car theft, etc.

In Washington, D.C. this trend was evident from the mid-1980s into the 2000s. Crime became prevalent throughout the 1980s, peaking in 1994, and then declining, which occurred across every major category of criminal behavior according to numbers from the Federal Bureau of Investigation’s (FBI) Universal Crime Reporting (UCR).

Property crime heavily correlated with violent crime (Pearson: .9183). Homicides heavily correlated with all violent crimes (Pearson: .9576).

Homicides Diverge Since 2012

But since 2012 the reported number of homicides in D.C. saw a random upsurge that no other major criminal category has seen.

Fourteen murders per 100,000 people in 2012 became 40 in 2021—a 111 percent increase or around 186 more murders in a year based on population estimates.

Yet rates of aggravated assault declined 31 percent over that time. Robberies declined 44 percent. All violent crime declined 35 percent.

As violent crime overall went down, homicides went up. The correlation of homicides with violent crime inverted, becoming a negative correlation (Pearson: -.6609). Effectively, homicide was no longer connected to other types of crime.

Connections to Declining Prosecution Rate

Oddities in D.C. crime data are nothing new.

Around the same time homicides started rising in 2010 is also when the U.S. Attorneys’ Office for D.C. (USAODC) began declining to prosecute a large number of crimes after jury trials regularly failed to convict.

The USAODC has continuously defended its prosecution record, stating that the cases they lost or declined to prosecute were simply bad cases and that the office lost 75 percent of search and seizure appellate cases because of changes to appeals law in recent years. The largest reason for cases being dropped was because of a lack of witnesses, not prosecutorial discretion, and there was little change in the number of convictions by guilty plea—not jury trials.

Consistency in Conviction Data

While data on conviction rates and dropped cases has all kinds of oddities, data on convictions shows that same consistent correlation across crimes—declining felonies parallels declining misdemeanors.

And the same goes for homicide convictions. Those also parallel the trends in overall felony convictions and misdemeanor convictions—a steady decline since 2010.

It’s as if the connection between homicides and other types still exists in conviction data, but not in incident data.