As the holiday season kicks into high gear, so do news reports warning us about crime. While the annual coverage has emerged as a wintry, anxiety-induced ritual, these reports are often based on anecdotes from police officials, or conventional wisdom proffered by security experts who don't offer any statistical support to back up their claims.
To find out whether there's any legit evidence of holiday crime trends, VICE spoke with Dr. Janet Lauritsen, a professor of criminology and criminal justice at the University of Missouri in St. Louis who's researched seasonal crime patterns. A report she co-authored for the Department of Justice (DOJ) last year found that property crimes were actually at their lowest during winter—apparently contradicting the annual bevy of bulletins from media outlets and police departments.
But the reality, it would appear, is a bit more complicated than that. Dr. Lauritsen dug deeper into the data to give us a fuller picture of the realities of holiday crime.
VICE: In your opinion, are news reports warning of holiday crime sprees warranted?
Dr. Janet Lauritsen: That's a good question. I'm familiar with the stories, so I [revisited] the data I use from the National Crime Victimization Survey (NCVS), which is an annual, ongoing survey where crime victims tell us what months their incidences occurred, regardless of whether they're reported to the police or not. If I [take] the last ten years of data and summarize the monthly totals for all different crime types, what I see is a December increase in only two types of crime, where the rate's a little bit higher than the average monthly rate, and that's for robbery and personal larceny—which are the crime types that police warnings concern. What I can see in this in data are patterns where January and February are extraordinarily low, but that December is a bit higher than average—about 20 percent than the average of all months combined.
So there is something to the warnings. I don't think it's necessarily about desperation, but I think it's more about opportunity—there's lot of people out shopping with goods on them. Personal larcenies are the crime that's about 22 percent higher in December than on average. Larceny includes the unlawful taking of property, other than cars, from the possession of another person by stealth, without force or deceit, [including] pick-pocketing, theft from motor vehicles, non-forcible purse snatching, and those kinds of things. Those do have a bit of a peak in December. I think it's an opportunity explanation more than a motivational explanation.
Your report also indicated that these types of property crimes are at their highest during summer. Is December's crime rate still lower than the rate of larcenies and robberies we're experiencing during summer months?
One thing that's interesting in the robbery data is that each season has something a little bit odd about it. For example, in the fall, October has a surprisingly high rate in the number of robberies compared to September and November. You can't make the motivational argument then that you can for Christmas—that people desperately want to steal Christmas presents for themselves or their families—because in October there's no holiday like that. But there is Halloween, and what Halloween does is bring a lot of people outside and partying. If there's going to be a [holiday] warning, we might want to offer one in October, too—a reminder that if you're out in large crowds, drinking with lots of other people in costume, that might be a likely time for robberies to occur, as there's more opportunities.
There are little October and September spikes in robberies, so it's not exactly a clean story. The seasonality story is generally clean; I'm comfortable saying there's a summer increase in burglaries, which are primarily committed by young people out of school.
With the larceny data, I see a higher point in December, but also in June and April. I can't think of stories for those months. This is the problem: when you try to spread crime reports across months, you're spreading the data thinner than you normally would. Even though it's a large sample, because victimization is a statistically rare event, you can sometimes see spikes in data that don't really mean anything, so you have to be cautious in interpreting monthly data. That's why in the report for DOJ we didn't analyze by month. There's just too much bouncing around in the monthly data to tell a clean story that can survive statistical tests. That's why we used seasons.
Two types of crime your report doesn't examine are crimes against retailers or cybercrime, which have also been claimed to occur more frequently during the holidays. Do you know of any research suggesting those crime rates are actually higher at this time of year?
No, I don't currently. I'm working on a larger project to modernize the nation's crime statistics with the National Academy of Sciences, and we're trying to pull together all the sources of data for different types of crime. And in terms of cybercrime, we really don't have the measures we need yet to make those types of claims with much certainty. However retailers do have inventory loss details, where they can pinpoint when their losses are more likely to be greater. The only source I'm aware of that might be able to get at that issue across the seasons is from the National Retail Federation. They'll release annualized estimates of inventory loss they've experienced, but I don't believe they release information on a monthly basis.
There are so many variations of cybercrime—there are crimes that are regular, like theft and fraud, but committed with the aid of cyber tools. The NCVS now has identity theft questions added to it, so there's one effort to get national estimates of identity theft. For things like cyberattacks, there are different consortiums of people trying to amass that data—business or technology groups that talk about millions of attempted attacks per second. We only hear about the breaches, but not the attempts, which are constant. I don't think there's any publicly available data there yet.
Overall, your report is pretty reassuring, as it shows the rates for all these crimes declining over time. Are there any findings from your research that you think would surprise people, as far as unusually high likelihoods of particular crimes during any season?
From that report the most surprising thing is, and I wrote a fuller academic paper on the issue, is what's going on with kids in the fall when they go back to school. Kids' period of risk is lowest in the summer, and that's because they're not near each other [to commit] simple assault, the most minor form of assault that kids are most likely to engage in. Summer is the period of least risk for kids.
When they go back to school in the fall, that's their period of highest risk, which decays over [the course of] the school year. It seems there's some process of challenging each other more so in the fall when they go back to school. Perhaps they're challenging disciplinary systems of their schools to see what they can get away with. Kids are the big drivers of victimization risk; younger age is correlated with victimization risk. They're responsible for a lot of that seasonality [in crime patterns], because kids are mostly the victims and also the offenders, too.
So ultimately for crime during the holiday season, you think all the normal tips that people receive, to draw less attention to their home and secure their belongings in public, are prudent advice to avoid becoming a victim of larceny or robbery?
Absolutely. These are just common sense reminders. There's no need to overdramatize it though. As soon as there's one horrible incident, most people would overestimate their risk.
This interview has been edited for length and clarity.
Follow Bill Kilby on Twitter.