How a Calif. agency’s theft deterrent strategies led to a 40 percent decline in auto burglaries
The Vallejo Police Department's crime prevention campaign aimed to reduce burglaries from automobiles during the holiday season
Many cops have heard of the SARA acronym, which grew out of the problem-oriented policing (POP) framework developed by law professor Herman Goldstein in the late 1970s. SARA stands for Scanning, Analysis, Response and Assessment.  
Problem-oriented policing emphasizes crime reduction through deterrence with the ultimate goal of being less dependent on the criminal justice system. POP, like evidence-based policing, works especially well when experienced police officers are partnered with competent crime analysts who are then able to “target, test and track” the data in a specific way that allows for strategies to be adjusted midstream if needed, with a rigorous evaluation to see what is working and, more importantly, what is not.  
The problem, however, is that in law enforcement, we often only scan and respond to problems, partly because many police agencies unfortunately do not have the budgets to employ crime analysts. The result is that we occasionally fail to analyze, and most of the time fall short of assessing the data. When we stop at only scanning and responding, we fail to drill down on the underlying causes of crime. In this vein, the Vallejo Police Department set out to use analysis and assessment to combat auto burglaries in a high-density shopping center while partnering with a crime analyst.
Attacking an increase in auto burglaries
Like many communities, auto burglary is a widespread problem in Vallejo, a city of approximately 122,000 located in the San Francisco Bay Area. According to FBI UCR stats for 2016, the city of Vallejo had one of the highest rates of burglary and larceny in California for a city its size. In 2016, we saw 983 incidents per 100,000. Although this is not as severe a problem as in neighboring San Francisco, which has the highest rate of any major California city, the problem was growing, with an 11% increase in 2016. 
During the 2017 holiday season, the Vallejo Police Department launched a crime prevention campaign titled HideitLockitTakeit, with the goal of reducing burglaries from automobiles. The campaign drew on the expertise of the department’s crime analyst and a partnership with BetaGov – a nonprofit research organization based at New York University that supports practitioner-led research. The campaign ran from November 8 through December 30 (holiday season) and resulted in a 40% decline in burglaries from autos compared to the same time period the previous year.
Using research to direct crime prevention efforts
Data help police zero in on where and when crime happens and who is committing it, which allows us to be strategic, rather than blindly implementing interventions or conducting enforcement.
In designing a campaign to address Vallejo’s property theft problem, we wanted to draw on as much existing evidence as possible. For example, research suggests that as much as half of all crime in any jurisdiction occurs at just 3-5% of locations.  Employing this knowledge, our campaign focused on a small geographic location in a high-density shopping center where theft from autos was occurring.
We also know that prevention involves reducing opportunities for crime or disincentivizing crime in an area.  Deterrence strategies include increasing the presence of officers and making threats of apprehension for specific offenses in an area. Such strategies have been shown to be more effective if they are limited in duration and rotated across different targets.  We designed our campaign accordingly by:
- Increasing officer presence by deploying unoccupied police cars and extra patrol;
- Leveraging police technology with the presence of GPS bait devices hidden in commonly stolen items – coupled with hidden cameras – and the deployment of WiFi hotspots for would-be criminals targeting computer devices left inside cars;
- Making implied threats of apprehension through our crime prevention flyers and electronic bulletin board posted on Hwy 80, and shared via local media outlets (newspapers, Facebook and Twitter);
- Increasing awareness of these crimes by providing tips to prevent them through “HideItLockItTakeIt” awareness flyers and by partnering with merchants;
- Focusing our efforts on targeted areas over a specific period of time and continually assessing our effectiveness with a crime analyst.
Other research-proven strategies to reduce property crime include reminding the public about their vulnerability to crime and increasing surveillance, particularly during times where crime rates are relatively high.
One study that focused on the theft of newspapers measured the effect of two types of messages. The first was a posted message that was politely stated and referred to rule-governed control. The second was more demanding and specified aversive consequences for theft. The study concluded that each sign was equally effective in reducing newspaper theft when compared to not having any notices at all.  Given this research, we created crime prevention flyers that provided awareness for citizens while placing potential thieves on notice that we were watching – not unlike many shoplifting signs.
Implementation of our crime-prevention campaign
The first component of the HideitLockitTakeit campaign was enforcement, in which we targeted thieves using our crime reduction team detectives, coupled with GPS bait tracking devices and undercover surveillance. We made a total of five arrests, all of which were “frequent flyers.” This enforcement phase lasted three weeks with the intention of rounding up those chronic offenders.
The enforcement phase was followed by deterrence interventions, which included focused patrol. We tested the impact of this deterrence intervention over the pilot period by randomizing the placement of empty police cars and the display of electronic bulletin billboards, coupled with overtime patrol shifts for 30 days while our crime analyst reviewed and assessed when and where thefts from autos were occurring.
The deterrence strategies continued with the assistance of electronic billboards, and our social and local media to promote the campaign, which helped to notify merchants and citizens by creating capable guardians through awareness in the spirit of situational crime control.  During December, we used additional focused patrols, crime prevention flyers and electronic billboards. Over the duration of the campaign, we continued covert and overt surveillance in the form of GPS bait technology, hidden cameras and unoccupied police cars – maintaining an enforcement component.
We discovered that some criminals were using personal iPhone hotspots to target and locate iPhones, iPads and MacBooks left in vehicles, often with their hotspots tethered to other devices. This act minimized the suspicion the criminals drew because it prevented them from having to look inside cars. We used local media to notify holiday shoppers of this criminal strategy.
Noting this strategy, we created a WiFi hotspot, called “Jason’s Macbook,” where we placed a portable WiFi hotspot inside our decoy car, which was coupled with a backpack or bag containing a commonly stolen item with GPS bait located inside – done in an attempt to attract thieves to our decoy car. We anticipated that this would lead to many arrests, but made only five arrests as a direct result of our GPS bait technology over the course of the 55-day trial. However, the GPS bait technology resulted in two search warrants where firearms were located and seized.
We believed the relatively small number of arrests occurred because would-be thieves initially had to be attracted to our parked undercover decoy car, which didn’t always happen. Later, the presence of our WiFi hotspots at times may have brought them to the decoy car containing our GPS bait, but we were unable to confirm that this was effective. The GPS bait technology intervention proved difficult. For example, one of our decoy packages contained a computer that was loaned to us by a local merchant and ultimately stolen by thieves that took our officers on a pursuit that we canceled due to the presence of the tracker and their reckless/unsafe driving. We trailed the subjects to Treasure Island in San Francisco where they located the GPS device and tossed it out the window. Unfortunately, the thieves were never identified, and the stolen computer was never recovered.
Increased police/citizen contacts
During additional patrols, some officers were naturally more inclined to make police/citizen contacts. Our data analysis and assessment showed a positive correlation to both fewer auto burglary incidents and more arrests on the days in which there were more police/citizen contacts. This result should not be a surprise, as research has long shown that increased police/citizen contacts help to suppress crime in localized areas.  
Blindly conducting increased police/citizen contacts without focusing on data trends can be limiting and even problematic toward community legitimacy efforts. A crime analyst can identify trends to guide increased focused patrols and theft deterrence strategies, making them more strategic than whack-a-mole strategies that may only displace crime. However, analysts enhance – but don’t replace – the value of instinctive police work that officers develop through years of experience working the street, sound relationship building, and even cultivating confidential street informants to tell us where and when a crime may be occurring.
A BetaGov statistician ran a correlation covariate analysis, which demonstrated that our deterrence methods were correlated with fewer auto burglaries (p<0.05). Most importantly, the analysis showed that police-citizen contacts were a strong predictor of positive outcomes and statistically significant (p<0.05).
The combined deterrence methods corresponded with a 40% drop for auto burglaries in 2017 from the same period in 2016. This is an important finding since the number of car burglaries in 2017 was greater than in 2016, leading up to this campaign. These significant findings highlight the need for a replication study, preferably using an experimental design to rule out seasonal/time-variant trends in crime.
Our 30-day focused patrol in December was done using officers on overtime, which is an expensive undertaking. However, a 2010 Rand study found that larceny costs an average of $2,139 per incident.  During our 55-day trial, we observed 17 fewer theft-from-auto incidents where we calculated an average loss of $1,368.30 per occurrence. As a result, the cost-benefit between both from the Rand study and our average loss was $36,364 to $23,261. The cost to the police department in overtime might be relatively minimal when compared to the potential saving in both tangible and intangible costs to the public from these theft incidents – but most importantly, the potential benefits of increasing our legitimacy in the community.
Our partnership with BetaGov helped us tailor our campaign. There are 18,000 police departments across the country, each with unique challenges, demographics and training – few have the ability to employ crime analysts. Nonetheless, we cannot take a one-size-fits-all approach. Policing is highly nuanced, complicated and involved; interventions must be tailored accordingly to each jurisdiction.  
Betagov also helped us disentangle various interventions to show which ones had a potential impact. When implementing multiple interventions simultaneously, there is always a challenge to determine causality. Many police researchers and evidence-based policing purists may prefer the police to implement one intervention at a time, to isolate the causal effect better. However, in the real world of policing – with real victims and a constant focus on reducing costs – it makes sense to implement multiple interventions simultaneously.
We showed that by continually evaluating data, we could have a measurable effect on crime in a localized area. Our data confirmed that police/citizen contacts were correlated to fewer auto burglaries and more arrests during the increased focused patrol. However, we also saw a decaying effect when we left the area as the rate of auto burglary returned, but for that holiday shopping season during the interventions, we lessened the chance of someone being a victim of auto burglary. There is no cure-all fix, but the hope is that by analyzing and assessing data, we can better understand the impact of police interventions, strategically inform our long-term operational decisions, and place cops at times and locations of crime to make us more effective in improving public safety.
1. Problem Oriented Policing (Center). What is POP?
2. Eck JE, Spelman W. Problem-solving: Problem-oriented policing in Newport News.
3. Sherman LW. The Rise of Evidence-Based Policing: Targeting, Testing, and Tracking. Crime and Justice, 2013; 42(1), 377–451.
4. Sernoffsky E. San Francisco’s auto break-in hot spots. San Francisco Chronicle.
5. Sherman L, Gratin P, Buerger M. Hot spots of predatory crime: Routine activities and the criminology of place. Criminology, 1989; 27(1), 27-56.
6. Clarke RV. Situational Crime Prevention. Crime and Justice, 1995; 19, 91-150.
7. Sherman LW. Police Crackdowns: Initial & Residual Deterrence. Crime and Justice,1990; 12 1-48.
8. Cordner G. "The Effects of Directed Patrol: A Natural Quasi‑Experiment in Pontiac," in James J. Fyfe, ed., Contemporary Issues in Law Enforcement. Beverly Hills, CA: Sage, pp. 37‑58.
9. Boydstun JE. San Diego Field Interrogation: Final Report. Washington, DC: Police Foundation, 1975.
10. Heaton P. Hidden in Plain Sight: What Cost-of-Crime Research Can Tell Us About Investing in Police. Santa Monica, CA: RAND Corporation.
11. Lum C. The Center for Evidence-Based Crime Policy - Introduction to Evidence-Based Policing.
12. Cambridge Center. A Conversation with Herman Goldstein. Cambridge Center for Evidence-Based Policing.