On the January 20, 2003, between the hours of 5:30 and 6:00 p.m., an armed private security officer (S/O) noticed a man shoplift some food items from the “Kash ‘n Karry” food store he was guarding in the Tampa, Florida, area. As the man left the store, the S/O pursued him and the suspect broke into a run. Blocks into the pursuit a strange thing occurred, the S/O accepted a ride from an unnamed motorist who wanted to participate in the chase. Coming upon the suspected shoplifter again, the S/O vacated the vehicle and confronted the suspected shoplifter. The suspect began to run again. The S/O had enough of the pursuit.  He went for his gun and fired a warning shot into the air from his duty-issued .38 caliber revolver. The alleged shoplifter dropped his loot and fled. The S/O retrieved the left-behind items and returned to the store. After the incident, the S/O was relieved of his duty gun, relinquished his security and gun licenses at the request of the state licensing officials, and the contract guard agency terminated the S/O for violating Florida law, licensing statutes, and his client’s wishes that S/Os not pursue shoplifters.

 

While incidents like the above are definitely unusual, occurrences where armed S/Os irresponsibly or mistakenly fire their weapons, unfortunately, may occur too often in Florida. These incidents help bolster the common perception that armed S/Os do not have enough experience and training to responsibly determine when to administer deadly force. In an effort to test whether or not that perception was justified, we studied reported licensed S/O firearm discharges in Florida from 2001 to 2008. Our study found perception might match reality more often than most would like to admit. When armed S/Os discharged their weapon in Florida, our study suggests that too many firearm discharges have been undertaken recklessly or accidentally, and sometimes contrary to the state’s use of force laws.

 

 

When armed S/Os discharge their weapon in the state of Florida, the incident must be reported. From 2001 to 2008, there were approximately 167 firearms discharge reports filed with the Florida Department of Agriculture and Consumer Services [1] (DACS), the agency responsible for regulating the state’s private security industry. Analyzing those incidents where an armed S/O discharged a weapon, we determined whether the discharge occurred due to a confrontation or an accident. We further broke down confrontations into security assignments across five mission types: retail security, residential patrol, bank security, armored car operations, and other. (We did not count incidents in the “other” category toward our study since they were infrequent. Incidents here included administrative violations by S/Os—such as wearing guns when not allowed by law—shooting at dogs or wildlife when fearing attack, and other minor incidents not involving people being hurt). Of the firearm discharges recorded, excluding the category of "other," about 50 percent were accidental or irresponsible, even reckless.

 

 

 

 

 

Retail Security. Armed S/Os working retail reported the most confrontations that ended with a firearm discharge, a little more than a third of all incidents. It was also the only sector that experienced an armed S/O’s death. These assignments involved armed S/Os patrolling locations like car lots, warehouses, department stores, supermarkets, nightclubs, and restaurants day and night and mostly in high crime areas. For the most part, firearm use resulted from S/Os being attacked immediately as the assailants entered the protected premises, S/Os walking onto a crime in progress, or S/Os responding to an altercation that suddenly escalated.

 

The DACS incident reports, however, did reveal an alarming habit. Approximately 10 armed S/Os assigned to food, liquor, and retail or department stores used warning shots to try and apprehend shoplifters despite DACS policy prohibiting such action. In addition, during some of the shoplifting incidents, S/Os fired at the car of the shoplifter as it attempted to leave the store parking area. S/Os stated they “feared for their safety” as rationale for firing at cars, but DACS reports detailed little factual descriptions of how S/Os were threatened. In one incident, the S/O’s stated rationale for firing at the car of a fleeing shoplifter was “he wanted to mark the car for police.” The firing of warning shots resulted in disciplinary action against S/Os about 80 percent of the time by DACS and criminal prosecution in some incidents.

Residential Security. Armed S/Os in residential areas experienced the second highest level of firearms discharges, about one in five incidents, during a confrontation. These assignments involved patrolling housing complexes by foot or car. Here armed S/Os, usually working alone, were outnumbered at least 2-to-1 in many “investigative” stops, where they question suspicious individuals. These stops also account for the most incidents where armed S/Os experienced edged weapons, like knives.

 

Even though residential patrol duties mean a lot of interactions with the public, and therefore more opportunities for conflict, most armed S/Os lacked intermediate and low-level force options—like pepper spray or a baton—that police rely on for managing physical encounters. This may be the reason that residential S/Os were the second highest group of S/Os that resorted to warning shots. Unlike retail settings, where S/Os tried to apprehend the perpetrator, residential S/Os seemed to use warning shots to dissuade individuals from physically attacking them, even if the attacker was unarmed. In one incident, an S/O responded to a domestic squabble in a parking lot by firing five warning shots from his pistol. The S/O was arrested for assault.

 

In addition, residential S/Os experienced the most “car shooting” incidents. Two issues surfaced here. Many car shootings were found justified because drivers—sometimes drunk—tried to hit S/Os with their cars. But S/Os also fired at vehicles without cause or by placing themselves in precarious positions, like blocking cars trying to leave the premises with their body, thereby bringing disciplinary measures or legal sanctions or both. In one incident, a teenager driving a car through a housing complex refused to stop when requested by S/Os. The guards, claiming the car was aimed at them, opened fire on the car, killing the teenager. Both S/Os were arrested and charged with murder. At trial, the S/Os were acquitted. Their security licenses, however, were revoked. Various civil lawsuits are still pending against the security agency and the residential complex over this incident.

 

Bank Security. S/Os assigned to bank security experienced the third most confrontation-related firearm discharges. All such firearm discharges occurred as a result of fighting off robberies and robbery attempts. Attacks here involved the almost exclusive use of handguns by attackers and ploys to distract S/Os.

 

In many cases, S/Os reported observing an individual enter the bank, look around, and leave – then return armed. Once again this shows the use of reconnaissance by assailants before an attack. In addition, these assignments also involved S/Os firing at moving cars. However, unlike the previously discussed incidents, here S/Os fired at moving cars to attempt the apprehension of armed robbers, although some incidents resulted in DACS discipline for public safety issues.

 

Armored Cars. Lastly, S/Os assigned to armored-car tasks experienced the fourth most confrontation related firearm discharges. Similar to bank security tasks, all confrontations here were related to robberies and robbery attempts.

 

However, it was not unusual for armed encounters here to involve sustained gunfire directed at S/Os, by multiple attackers, using handguns and long arms. A common theme in most armored car robberies was the use of well-timed teamwork to distract S/Os, during pickups or deliveries, indicating the use of intelligence gathering by criminals to gain an edge by using surprise as a weapon. Some surprise ambushes were so effective that S/Os were disarmed and money taken without shots fired.

 

But some attacks simply employed brutal tactics—showing a critical challenge these assignments face. For example, in one attack, a van screeched to a halt in front of an armored truck as an S/O filled an ATM machine. Two men wearing ski masks and carrying AK-47 rifles exited the van and opened fire on the S/O at the ATM machine and the S/O sitting in the armored truck. Both S/Os were shot but managed to fight off the attack. One bullet penetrated the armored car and hit the driver. Despite the level of violence demonstrated by such attacks, armored car assignments resulted in very low casualties. Interestingly, no S/O assigned to armored car operations fired warning shots during the eight years this study focused on.

 

 

By breaking down the discharges by assignment, we did find some trends. In short, almost all reported confrontations occurred in major urban cities. Second, most S/Os seemed to rely on handguns as the main force option even when the aggressor was unarmed. Third, in all assignments, S/Os fired at moving cars – no DACS policy or Florida S/O training addresses this issue.

 

Lastly, some confrontations related to specific assignments seemed to differ by region. For example, half of all armored-car-related gunfights occurred in Miami. The Tampa-Ft. Myers area was the locale for the most retail-security confrontations. The Orlando “metro” area experienced the highest number of confrontations related to residential security, but only by 4 percent. Residential armed confrontations were distributed almost equally across the state, showing a unique training need. Approximately 90 percent of all armed confrontations involving bank security occurred in South Florida, split equally between Miami and West Palm Beach. Eight years of data here suggest security operations planning efforts, including liability mitigation efforts for security operations, must adapt and account for regional threat trends that state and other regulatory training standards might not account for. For instance, Miami armored-car guards may require additional firearms training.

 

 

 

To further understand what S/Os faced during confrontations we scrutinized DACS reports to discover any other data about S/O armed confrontations that may suggest why the security guard made the decision to fire. We found the following:

 

Gun Choices. While many Florida police agencies carrying “.40” caliber semi-automatic pistols, DACS policy restricts security weapons to 9mm semi-auto pistols and.38 caliber revolvers and the 12 gauge shotgun—although a waiver is available upon request. DACS incident reports suggested that over half of the firearms used in incidents were .38 caliber revolvers. While selecting a higher caliber pistol or semi-auto pistol is not critical, since the major police agencies still use 9mm, our findings suggest that the security industry has not “updated” any training or studies on handgun choice. For example, our own law enforcement experiences and visits to police departments nationwide lead us to conclude that most, if not all, police agencies have adopted the semi-auto pistol for its inherent tactical advantages: it fires more rounds, reloads faster, and is less bulky and easier to shoot than the revolvers many security guards carry.

 

 

Distance. As in law enforcement, S/O armed encounters occurred at very close distances—we estimate over 90 percent occurred under 15 feet, probably closer, from description of events. However, armored car and bank security confrontations involved longer distances on the average.

 

Body Armor. Most S/Os did not wear body armor. Our police experience leads us to recommend body armor, which can help during knife and physical attacks as well as during other accidents. Body armor is not addressed by DACS policy. But the protection does matter, according to the incident reports. One S/O not wearing body armor was killed when he was shot in the chest during a retail assignment, but another S/O survived being shot in the chest during a residential assignment because he was wearing body armor.

 

Tactics. Reports disclosed that many S/Os approached situations without thinking about cover, without backup, or calling “911”—despite having the opportunity to do so. Furthermore, they did not employ “contact and cover” strategies when more than one S/O was involved. Also, S/Os generally did not verbally warn a perpetrator before escalating to warning shots. However, there were S/Os that performed admirably, escalating situations as dictated by the attacker.

 

Low-light Scenarios. In low-light scenarios, S/Os sometimes assumed the attacker was armed and this led to the S/O firing their handgun. Flashlight use was not reported to help with low-light conditions. DACS mandated training does not address low-light training. Most modern police forces mandate low-light training and qualification drills.

 

Hit Rates.  S/Os managed a 14 percent “shots fired versus hits rate” from 2001 – 2008, but the rate is severely skewed. Annually, the rate was dismal: In 2008, the hit rate was 2.27 percent. In 2007, the hit rate was near zero percent. Two years earlier, however, two officers did remarkably well - one going 6 for 6, bringing the hit rate to above 28 percent. If not for these S/Os, the hit rate would be under 1 percent for that year and would have lowered the 8-year-study hit rate.

 

In comparison, the New York City Police Department (NYPD) achieved a 28.3 percent hit rate in 2006. In Los Angeles, the police achieved a hit rate of 40 percent in 2006. In Baltimore, MD, police achieved hit rates of above 60 percent in daylight hours and above 40 percent during low-light encounters, due to training initiatives there. Based on the data, shots by S/Os seemed likely the result of instinctive point and shoot reactions. Surprisingly, hit rates were similar with revolvers or pistols, day or night, which is unusual in our experience.

 

Rounds Fired. S/Os with pistols fired approximately 3-6 rounds, those with revolvers, usually 1-2. This is somewhat similar to police. Before the NYPD adopted pistols, the average number of shots fired per engagement was about 3.6 rounds. After pistols were adopted, this rate went up to about 4-5 rounds. In Portland, Oregon, on average, shots fired by cops per incident went from 2.6 to 4.6 shots fired after adopting pistols. However, unlike S/Os, the hit rate for police using pistols has gotten better because of training. The necessity for rapid reloading to prevent death/injury was not a factor in any of the cases examined. Most interactions were over in seconds.

 

In addition, since many S/Os in Florida still carry revolvers, an interesting comparison was possible. Based on data reviewed, on average, S/Os with revolvers fired less rounds than S/Os with pistols, during similar incidents, yet were just as effective in stopping or scaring away the attacker. And for reasons unknown, S/Os assigned to retail security assignments fired three times the amount of rounds overall than S/Os assigned to any other assignment. Nevertheless, despite police hit rates rising due to the use of semi automatic pistols. S/O hit rates in this study showed no improvement since the adoption of semi-automatic pistols. This may denote a need for training changes which we discuss below.

 

Other. The data collected suggested that S/Os drew their gun mostly to confront other human beings—as opposed to shooting at dogs, which occurred several times. This is underscored repeatedly by adversaries’ “willingness” to challenge S/Os using physical force during retail and residential security assignments—including attempts to disarm S/Os. DACS policy does not require weapons retention training or defensive tactics or physical fitness training.

 

 

 

The most disconcerting statistic revealed by our analysis is the frequency an armed S/O’s firearm “just went off.” Accidents, defined as the negligent discharge of a firearm, accounted for over a third of all discharges, which strongly suggests a weakness in firearms training.

 

Data collected showed that S/Os accidentally fired their guns 38 percent of the time from 2001 to 2008. However, viewed annually, this matter was more pronounced. First, we found that reported accident discharges sometimes outnumbered reported confrontation discharges annually. In our experiences, such data suggests deficiencies in training, discipline, and oversight–a serious liability matter. Second, that there was a high rate of accidents involving revolvers—which are a simpler kind of firearm than a semi-auto pistol—gives credence to our opinion of a training deficiency. Third, data suggested that close to half of the accidents occurred during the loading/unloading process—whether using a revolver or a pistol – a training matter. Fourth, some accidents occurred during “operational tasks,” such as drawing the gun to have it ready as directed by policy.

 

 

For example, in all three accidents involving shotguns deployed during armored car “docking” operations, S/Os slipped, tripped, or stumbled while carrying the shotgun “at the ready,” with their finger on the trigger and the safety off—which is not done in policing. More perplexing, however, was that about 50 percent of the accidents occurred when S/Os were bored and decided to check their gun, clean their gun, move their gun belt around, dry-fire, show off their gun, take the gun apart, practice quick draws, or unload or load their gun while on post—a potentially dangerous habit. In one incident, a supervisor directed various S/Os to dry-fire due to “down time,” resulting in an accident. All were disciplined by DACS. Interestingly, most accidents were explained by S/Os as the gun firing itself—despite evidence to the contrary.

 

More worrisome, DACS reports showed that some accidents were disclosed only after supervisors found bullet holes in cafeteria refrigerators, employee room walls, bathroom stalls, office computers, post surroundings, and even a bleeding dog walking around a protected housing complex. Further inquiry often led to S/Os confessing that they did not report the accident for fear of losing their security and gun licenses as well as their job. In other incidents, it was documented that the S/O begged witnesses not to report the matter.

 

For comparison purposes, the following are some of the statistics we found regarding law enforcement “accidents.” Armed S/Os do not fare well. First, the Contra Costa Sheriff’s Office in California, which has 800 deputies, reported no accidental discharges in over 5 years in 2006. Additionally, a study of the New York City Police Department (NYPD), which has over 32,000 officers, who have constant access to their firearms, unlike S/Os, reported 24 accidents in 2005 and 26 accidents in 2006. In our police experience, on-duty accidents involving firearms are usually the result of weak training and lack of discipline and supervision.  

 

 

While police agencies, in general, have made remarkable progress in updating firearms training, private security has not—unless regulatory mandates were enacted that is. In New York, for example, 47 hours of peace-officer-firearms training was mandated for all armed security officers due to public safety concerns in 1997. The New York Legislature regulated training because of concerns that the security industry would not do it unless required to. That skepticism seems warranted. Gaining management support for line personnel training can sometimes be a major financial challenge in the private security industry, according to armored car companies we spoke to who requested anonymity. Nevertheless, more intensive, and therefore expensive, training is a necessity for armed S/Os.

 

After reviewing the shooting reports discussed herein, we found the following differences between private and public sector training in Florida for carrying a firearm in public as a duty:

 

 

Based on the above, S/O training is far less demanding than police training. Two problems therefore face the private security sector as it moves to fill more aggressive roles post-9/11. First, the differing training and licensing standards among the 50 states only serve to promote conflicts that allow litigators to pit one standard against the other, confuse trainers and S/Os, and mislead employers about risk management for various assignments. For instance, some residential security posts may require training in handling domestic disputes as first responders. Second, armed S/Os have the responsibility to judiciously exert authority over the liberty and life of the public at large. Basic negligence and tort principles demand comparable skills to other armed professionals plying their trade in public, especially when managing similar risks.

 

Analyzing eight years of armed S/O discharge reports demonstrate that public safety demands training tailored to the needs of the position, not meet minimally required standards. The more and more S/Os are asked to perform tasks associated with traditional policing, the more and more they need to be trained to faithfully execute their duty to protect people and property. The security industry should take the lead and prepare its line personnel for the more complex work that’s in demand.

 

Javier Soto, JD, is a retired federal agent and former police firearms instructor, and a graduate of various federal, military and local police SWAT schools. He also served in internal audit/internal affairs type roles in two different federal agencies. Reginald Kornegay, director of Executive Preparedness Group, LLC, is an experienced senior level federal regional emergency management official and trainer and a senior level military officer (reserves). He is also a former federal agent and police officer and has completed numerous SWAT, sniper, and other tactical programs. Kornegay has developed and provided leadership, emergency management, and line officer training for domestic and foreign police and military forces.