OP-ED | Crime Statistics Aren’t as Telling as You Might Think
The FBI released its crime statistics for 2016 last week, and there has already been a lot of discussion and commentary about the increases in murder and violent crime rates for a second year in a row.
To put things in perspective, some more thoughtful reviews of data point to the 20- to 25-year streak of declining crime rates and the fact that a one or two-year slight uptick does not mean much.
To add even more perspective on the issue, it is important to consider how crime rates are calculated. In a sense, they are like an “average body temperature among all patients in a hospital” — some have higher temperatures, some have lower, and an average might not be that useful as an indicator. Generally, only a few cities (the biggest ones) contribute to the bulk of crimes reported to the FBI and thus disproportionately affect the overall crime rates in the U.S.
Considering this fact, most experts point out that crime is a local phenomenon and national averages should be interpreted with caution. You undoubtedly have knowledge of which neighborhoods are considered safe in your city or town and surrounding areas. In almost any city, some neighborhoods bear the brunt of crime problems like shootings and violence while other neighborhoods are relatively quiet. And the distribution of crime among local areas does not change much year to year.
Among Connecticut’s 90 cities and towns whose police departments have submitted their crime data to the FBI, more than half (50 cities and towns) have extremely low rates of violent crime: less than one violent crime a year per 1,000 residents. For these places, about 97 percent of all index crimes reported to police are property crimes. So, keeping this skewed ratio in mind, let’s consider in more detail what goes into calculating “violent crimes” and “property crimes.”
Violent index crimes are murder (and non-negligent manslaughter), rape, robbery, and aggravated assault. The most common among these crime types is aggravated assault and thus, it contributes disproportionately to the violent crime numbers and rates.
The second most numerous category is robberies. Together, aggravated assaults and robberies comprise 90 percent of all violent crimes reported in Connecticut. Thus, rapes and murders (arguably, the most violent crimes) contribute very little to the “violent crime rate.”
At the same time, the ratio of contributing crimes varies by location. Almost half of all reported violent crimes in Connecticut (48 percent) occurred in three large cities: Hartford, New Haven, and Bridgeport. And 94 percent of their reported violent index crimes are robberies and aggravated assaults. The same skewed ratio was true for New London and Waterbury. Among the rest of the state’s cities and towns, the ratios varied quite a bit, with West Hartford taking a prize on the most unusual mix for year 2016: zero assaults but 38 robberies (to constitute 95 percent of West Hartford’s violent crimes).
Now let’s consider what goes into property crime numbers. Index property crimes are burglary, larceny/theft, and motor vehicle theft. In addition, the mix is actually even more skewed than the violent crime compilation. Among Connecticut’s 60,000+ property crimes reported last year, one category clearly overwhelms the rest: larceny/theft contributes almost three quarters (about 74 percent) of all property crimes reported to police. As for the mixes of property crimes among various cities and towns in Connecticut (including all places with more than 1,000 residents), variations are not huge. This time, a prize goes to Easton and Weston for having zero motor vehicle thefts; with Old Saybrook, Granby, and Redding taking the second place, with one motor vehicle theft reported in each town in 2016.
As for property crime rates overall, they do vary quite a bit among different places in the state. Three Connecticut cities with the highest rates of property crimes are New Haven, Hartford, and Waterbury. However, unlike with violent crime numbers, these three cities together only contributed less than a quarter of all property crimes in the state (24.8 percent). Their property crime rates hover around 40 crimes per 1,000 residents for 2016, and a large majority of the state’s cities and towns (62 out of 90) are not dramatically far behind, with property crime rates ranging between 10 and 30 crimes per 1,000 residents.
Only about two dozen towns (mostly smaller ones, plus Greenwich) have property crime rates below 10 per 1,000 residents, with Weston, Ridgefield, and Newtown having the lowest property crime rates in 2016 (ranging from 1.4 to 3 per 1,000 residents).
Yet to put the property crime rates into perspective: index property crimes do not take into account any of the online property crimes (like most identity and credit card fraud). Last year, my colleagues and I published a journal article demonstrating that a lot of the property crimes might be moving into the online sphere and thus do not contribute to the crime statistics reported by the FBI (instead contributing to what is called the “dark figure of crime,” or unreported crime).
Thus, there is little meaning to the claim that there is a “falling property crime rate” in Connecticut, and in the United States as a whole. Based on our analyses of available data from other sources, we have concluded that cybercrime and identity theft are on the rise, despite the continuing decreases from year to year in the “traditional” reported property crime statistics.
Unfortunately, the crime statistics the FBI releases are essentially, old news. Unlike the key economic indicators that are calculated almost in real time (for example, monthly updates on jobs and unemployment figures are released just a few days after each month ends), crime statistics are notoriously slow and cumbersome to collect.
Many police departments do not report their data to the FBI or report partial (and hence often misleading) data, and there is no easy way to remedy the situation. Therefore, we are now analyzing and fretting over last year’s data, especially when we are looking at the violent crime upticks.
Although U.S. Attorney General Jeff Sessions issued a dire warning about the “rising tide of violent crime” in the United States, this year’s crime stats, in all likelihood, are better than last year’s. According to Darrel Stephens, an executive director of the Major Cities Chiefs Association, as well as an analysis by the Brennan Center for Justice, current year crime statistics look much better than the previous year’s. Both the violent crime rate and the murder rate are expected to decline in 2017.
In conclusion, we should exercise caution and a healthy dose of skepticism when interpreting the FBI Crime in the United States data. To be fair, the FBI does continuous work on improving the data quality and usefulness of their statistics. For example, they have discontinued the use of the Crime Index (that used to lump together violent and property crimes and thus was mostly reflective of the most numerous category of crimes: larceny/theft). Another welcome development is that the FBI has discontinued the use of one of the most misleading and alarmist ways to represent crime: the Crime Clock.
Yet, we should still be careful with the violent crime index and property crime index. Finally, we should keep in mind that the so-called “street” property crimes do not reflect the large and likely growing category of financial cybercrimes.
Maria Tcherni-Buzzeo is an associate professor of criminal justice at the Henry C. Lee College of Criminal Justice at the University of New Haven.
DISCLAIMER: The views, opinions, positions, or strategies expressed by the author are theirs alone, and do not necessarily reflect the views, opinions, or positions of CTNewsJunkie.com.