Safe in America — what the FBI crime data really tell us

How safe is America? Don’t expect the complete answer when the FBI releases the 2016 data from the national Uniformed Crime Reporting program this month. These crime statistics provide a barometer of reported crime; one measure of safety for the country.

They also provide rhetorical opportunities — on both sides of the aisle — for trumpeting success when crime is down, or beating the drum for new policies and more resources when it is up. However, recent history has taught us that data on crime is vulnerable to misinterpretation and that inaccurate stories about spiking crime rates can quickly take on a life of their own.

{mosads}Two years ago, a number of media outlets reported that early crime estimates in a few cities indicated the beginning of a new wave of violence, invoking images of the high crime rates that we saw across the country during the 80s and 90s. This news led to a series of alarmist headlines that we were facing a surge in violent crime.

With hindsight, we now know these conclusions were premature, and 2015 ended close to a 50-year low for homicides — with 4.9 homicides per 100,000 population, compared to 10.2 per 100,000 in 1980 and 9.8 per 100,000 in 1991.

In a Vera Institute of Justice’s recent report, Measuring Public Safety: Responsibly Interpreting Statistics on Violent Crime, we analyzed common pitfalls and lessons learned when interpreting crime data — lessons that we expect will apply as well to the 2016 crime statistics.

Lesson No. 1: While it is tempting to interpret an increase in crime from one year to the next as the beginning of a crime wave, even if we see an uptick for some crimes or in some places, a year-over-year increase does not constitute a trend. In our brief, we examined data for 10 cities that were highlighted in 2015 as being at the vanguard of spiking homicide rates, among them Chicago, New Orleans, New York City, Dallas and Washington, D.C.

More recent data requested from police departments in each of those places indicates that, for the three of the 10, an increase in 2015 was followed by a decline in homicides in 2016. Three went on to experience a sustained increase in homicides that were unusual and the remaining four ended 2016 at or near a 40-year low point for homicides.

Lesson No. 2: All crime is local — and so are most solutions — something that easily gets lost when interpreting national data on crime and violence. In other words, the factors that impact crime play out differently from place to place and responses to crime should be tailored to these local realities.

Some fluctuation is inevitable when looking at statistical data and when the new crime statistics are released for jurisdictions across the country, we will undoubtedly see that crime is up in some places and down in others. Furthermore, some places that have experienced rising crime in recent years may continue to become less safe.

In recent years, Chicago, Baltimore, St. Louis and Milwaukee have all seen fairly dramatic increases in homicide rates that demand our attention. However, even within these cities, violence is often hyper-localized. In Chicago, for example, 20 percent of homicides reported in 2015 occurred in just two of the city’s 25 police districts. We should focus resources on these places and apply what we have seen works to reduce crime and save lives.

What we should not do is interpret the increases in these cities as a “nationwide” problem. The national homicide rate fell by 55 percent between 1991 and 2014. While there was a slight increase in 2015, the rate remains at half the 1991 level. Furthermore, 20 percent of the increase in 2015 was in four cities — Baltimore, Chicago, Milwaukee and Washington. There are abundant examples of places that have successfully addressed violent crime. Accounting for changes in the population, there were more than seven times as many homicides in New York City in  1990 compared to 2015 and a number of other cities including Dallas, Boston and Los Angeles saw similarly dramatic improvements in public safety over the same period.

It is clear we have an opportunity to learn from recent history and we should exercise caution before reaching broad conclusions. It would be a mistake to upend the national justice polices of recent years, based on a few neighborhoods in a subset of cities that are experiencing serious problems with crime and violence. Political scaremongering and kneejerk responses to short run crime statistics run the risk of undoing the progress that we have made to date, and will distract from the focused, localized work that still needs to be done.

If we can put rhetoric aside, early returns show that — with some important exceptions — we continue to live in the safest period in more than a generation.

Jim Parsons is vice president and research director at the Vera Institute of Justice.