The topline for most press coverage of yesterday’s release of the FBI’s 2020 Uniform Crime Reporting (UCR) program data is – unsurprisingly – that America has become a more violent place. The violent crime rate rose by 5.2 percent over 2019 figures, driven by a 29.4 percent increase in overall murders and a 12.2 percent jump in aggravated assaults. Violent crime was already elevated in the first few months of 2020 and increased significantly from June until the end of the year. And violence was worse everywhere, regardless of region or population size. Notably, 2020 logged the largest share of murders committed with a handgun (77 percent) since record-keeping began.
That’s the bad news. And it is very bad news indeed, especially for the communities where violence is entrenched, cyclical, and seemingly unresponsive to law enforcement’s best efforts. But there are considerable silver linings in the data as well. The rates of robberies, rapes, burglaries, and thefts dropped significantly, contributing to an almost 5 percent decline in overall major crimes. Some of this drop was likely pandemic-related: fewer opportunities for crime meant fewer crimes committed. Still, property crimes fell (by 7.8 percent) for the 18th consecutive year, which suggests that the trend will outlast COVID.
Still, there are other less obvious things to worry about beyond the murder rate. The arrest rate for violent crimes stayed roughly the same year-over-year in 2020, despite more incidents of violence, and the arrest rate for property crimes fell by about 22 percent, outpacing the drop in crimes. These data suggest a troubling trend continued in 2020: a decline in the clearance rate (that is, the rate at which crimes are solved). In 2019, only 45.5 percent of violent crimes and 17.2 percent of property crimes were cleared. 2020 looks to be even worse. The UCR data also show that police departments, particularly those in large cities, were understaffed, which undoubtedly contributes to the clearance problem.
The biggest problem with the UCR has nothing to do with this year’s data. Rather, it has to do with the foundational assumption of the program: that enough law enforcement agencies will share their numbers with the federal government. The FBI received information from 15,875 out of 18,623 agencies in 2020, or about 85 percent. That’s lower than the rate of participation was 30 years ago, but enough to paint a representative picture of the scope and nature of crime in America. But the UCR will soon be phased out in favor of the more detailed National Incident-Based Reporting System (NIBRS), which has many advantages over the UCR, though robustness is not one of them. Only 9,991 agencies submitted data to NIBRS last year, covering about 53 percent of the U.S. population. That submission rate isn’t even adequate for the FBI to generate quarterly reports, much less accurate year-end statistics.
Part of the problem is that the incentive for participation in federal crime reporting programs is eligibility for federal grants, and many – even most – law enforcement agencies are too small to bother with them. Another issue is the standard to which data are held: many agencies make good faith efforts to submit data that the FBI then rejects. The upshot is that NIBRS is stuck with a participation rate that the UCR surpassed in the 1950s.
Researchers rightly worry that low reporting and the shift to NIBRS could be catastrophic to their efforts to accurately describe crime in the U.S. Rising rates of violent crimes are scary and worthy of notice. But we may reach a point soon when we cannot tell with any certainty what the violent crime rate is, much less in what direction it’s headed. And that’s a big problem because what gets mismeasured will ultimately get mismanaged.